Commit Graph
8055 Commits
Author SHA1 Message Date
DottaandClaude Opus 5.5 6b6c94a4ea perf(chat): read the GitHub response run by issue run indexes (#21577)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The GitHub chat channel keeps one response comment for each GitHub
request. Two background sweeps (acknowledgements and terminal responses)
and the acknowledgement step read the latest run of the assigned agent
for the request's issue with `readGitHubResponseRun`
> - That function sends two statements. Both match runs with
`coalesce(native_issue_id::text, context_snapshot->>'issueId') = <issue
id>`. No index has that expression, so each statement reads every run of
the agent and decompresses the large `context_snapshot` of each legacy
run
> - On a hosted test database, these two statements were the two most
expensive statements by database time. In a 701 s window, each ran 6.7
times a second, for 203 s and 137 s of database time (about 30 ms and 20
ms for each call). Together they kept about half of one database session
busy
> - This pull request reads the latest run with two index lookups
(native binding, and legacy context without a native binding), the same
way #20671 changed the terminal-response sweep. It decides request
ownership on that one run in the same statement
> - The benefit is that each call reads a few index entries in one
statement instead of the agent's whole run history in two statements.
The results do not change, and no migration is necessary

## Linked Issues or Issue Description

Refs #20671 (the same rewrite for the terminal-response discovery page;
its Risks section names this function as the remaining case).

**What existing behavior does this improve?**
`readGitHubResponseRun` in
`server/src/services/chat-channels/github/response-run-state.ts`. The
GitHub acknowledgement sweep calls it once for each candidate request.
The terminal-response sweep and the acknowledgement step call it through
`githubResponseExecution`.

**Subsystem affected**
GitHub chat channel: request acknowledgement and terminal-response
recovery.

**Current behavior**
The function sends two statements:

1. The latest run of the agent for the issue that woke for the request:
`heartbeat_runs` joined to the inbound `chat_message_links` row of the
delivery on `wakeCommentId` or `wakeCommentIds`, ordered by `created_at
desc`, limit 1.
2. The latest run of the agent for the issue, with the
`responseRunColumns` projection, ordered by `created_at desc, id desc`,
limit 1.

Both statements filter on `coalesce(native_issue_id::text,
context_snapshot->>'issueId') = $issueId`. The planner can only use
`heartbeat_runs_company_agent_started_idx` or
`heartbeat_runs_company_created_at_desc_idx`. It reads all runs of the
agent and evaluates the `coalesce` on each row, which decompresses the
context of each legacy run. On the hosted test database,
`heartbeat_runs` has about 31,000 rows and 1.96 GB, because
`context_snapshot` is large.

**Proposed behavior**
One statement:

- The latest run ID comes from two disjoint branches, each `order by
created_at desc, id desc limit 1`: runs with `native_issue_id =
$issueId` (through
`heartbeat_runs_company_native_issue_contract_id_uq`), and runs with
`native_issue_id is null and context_snapshot->>'issueId' = $issueId`
(through `heartbeat_runs_company_ctx_issue_created_idx`). Together the
branches are exactly the rows where the old `coalesce` equals the issue
ID.
- The statement reads that one run by primary key with the same
`responseRunColumns` projection.
- An `exists` subquery on the inbound `chat_message_links` row of the
delivery decides whether that run woke for the request. It uses the same
`wakeCommentId` / `wakeCommentIds` match as before, but on one run only.
- When the issue ID is null, the function returns no run without a
query. The old statements could not match a null issue ID either.

**Reason and benefit**
These were the two most expensive statements on the hosted test
database. On a local fixture with the hosted shape, the median call goes
from 111.6 ms to 1.03 ms (about 108 times faster), and the buffers for
one call go from about 73,000 to 98.

**Breaking changes**
None. The function signature, the returned run, and the ownership result
do not change.

## What Changed

- `readGitHubResponseRun` reads the latest run of the agent for the
issue through the two index-backed branches, and decides request
ownership on that run in the same statement. Two statements become one.
- Add `response-run-state.test.ts`. It runs the old two-statement
function (kept in the test as the oracle) and the new function on the
same inputs, and it requires equal results. It also requires a plan
bound: one call must read fewer than 150 `heartbeat_runs` rows while the
agent has 1,500 unrelated runs.
- Add `scripts/benchmark-github-response-run-state.ts`. It loads the
baseline module from a given commit, compares the complete output of
both versions for every input, measures both, and prints the plans.

## Verification

Why the old statements were slow, and the thesis: each statement filters
the agent's whole run history with an expression that no index has, and
detoasts the context of every legacy run. If each precedence branch uses
its own issue index and ownership is checked on the one latest run, the
cost no longer depends on the agent's run count, and the results stay
identical.

Red/green regression test:

- `cd server && npx vitest run
src/services/chat-channels/github/response-run-state.test.ts`
- Red (test file on the old `response-run-state.ts`): the plan bound
fails with `expected 3597 to be less than 150`. The equivalence test
passes, because the oracle is the old code.
- Green (this change): 2 of 2 pass.

The equivalence fixture has two companies, two agents each, 60 issues
each, and one or two GitHub requests for each issue. Runs cover these
cases: native binding, legacy context without a native binding, a native
binding to another issue while the legacy context names this issue,
another agent, another company, `wakeCommentId`, `wakeCommentIds` as an
array, `wakeCommentIds` as a JSON string, a link to the outbound message
of the delivery, a link to another company's comment, a link to the
sibling request of the same issue, a request with no inbound comment,
ties on `created_at`, and a null issue ID. It also calls with another
company's endpoint. The test requires more than 20 results in each
outcome (owned, latest run not owned, no run) and both binding kinds.

Local benchmark with the hosted shape:

```sh
pnpm --filter @paperclipai/server exec tsx ../scripts/benchmark-github-response-run-state.ts --baseline-ref=3c3e6c2d75
```

Fixture (synthetic): one company, 4 agents with about 7,750 runs each
(31,000 runs), 2,000 issues, 300 GitHub requests. Each run has about 60
KiB of incompressible context, and `heartbeat_runs` is 1,932 MB. Runs
are native or legacy at random, and about half link to the request
comment. Embedded PostgreSQL 18.1 on an Apple laptop. The hosted
database is Neon, so absolute times are local measurements only. The
benchmark compares the complete output for all 600 inputs (262 owned,
329 latest run not owned, 9 no run): equal. It then times 20 inputs with
runs (two agents), with 2 warmups and 7 alternating samples.

| | Statements per call | Median ms per call (range) | Buffers for one
call |
| --- | --- | --- | --- |
| Before | 2 | 111.6 (100.9–240.5) | 18,358 hit + 26,498 read, then
14,997 hit + 13,489 read |
| After | 1 | 1.03 (0.81–1.39) | 80 hit + 18 read |

Plans for one call. Before: statement 1 is a `Bitmap Heap Scan` on
`heartbeat_runs_company_agent_started_idx` that removes 7,674 rows by
filter (75.7 ms). Statement 2 is an `Index Scan` on
`heartbeat_runs_company_created_at_desc_idx` that removes 19,801 rows by
filter (60.2 ms). After: one `Index Scan` on `heartbeat_runs_pkey`. Its
`InitPlan` is a `Merge Append` of two `Limit` index scans on
`heartbeat_runs_company_native_issue_contract_id_uq` and
`heartbeat_runs_company_ctx_issue_created_idx`, and its `SubPlan` is one
`Index Scan` on `chat_message_links_delivery_idx` (0.128 ms).

Other checks:

- `cd server && npx vitest run
src/__tests__/chat-channels.integration.test.ts -t ithub`: 104 passed
(all GitHub chat-channel integration tests, including the
acknowledgement and terminal-response sweeps).
- `cd server && npx tsc --noEmit`: passed.
- `pnpm verify` exited 2 because a file under `scripts/` changed, which
requires the broad CI coverage. Required CI covers the rest.

Why the function runs 6.7 times a second. The acknowledgement sweep and
the terminal-response sweep rebuild their candidate set when a hint
arrives, and they call this function once for each candidate on each
rebuild. A run milestone for a run with no issue (for example, every
status change of a run without a task in `setRunStatus`, or a comment
merged into an active run in the wake queue) sends a company-wide hint
to both sweeps. So in a busy company, both sweeps rebuild about once a
second (their minimum pass interval), and each rebuild reads every
parked candidate again, although nothing about those candidates changed.
In a local reproduction with 8 parked candidates (4 for each sweep) and
the real coordinator and workers, 10 company-wide milestone hints a
second gave 8.75 calls a second; 1 hint a second gave 7.98 calls a
second; 10 hints a second for an unrelated issue gave 0 calls; and no
hints gave 0 calls. So the hosted rate is consistent with about 7 parked
GitHub requests in one company and a steady stream of company-wide run
milestones. This pull request makes each call cheap. A separate pull
request will address the repeated rebuilds.

## Risks

- Low risk. The change is one function. The run match is the same set of
runs: `native_issue_id::text = <uuid text>` is equal to `native_issue_id
= <uuid>` for the canonical UUID text that callers pass (the issue IDs
come from UUID columns), and a null `native_issue_id` makes the old
`coalesce` use the legacy context.
- Ownership: the old code compared the latest run that woke for the
request (ordered by `created_at` only) with the latest run (ordered by
`created_at, id`). That is true exactly when the latest run woke for the
request, except when two runs that both woke for the request tie on
`created_at` with the latest run. In that case the old result depended
on the physical row order. The new code always returns the deterministic
answer (the latest run woke for the request).
- The number of calls does not change in this pull request.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic. Used through Claude
Code with tool use (shell, file edits, and local PostgreSQL benchmarks).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 00:59:25 -06:00
Nicky LeachandClaude Opus 5.5 e977c3b922 feat(native-runtime): end the wait for the provider's own turn when a person comments (#21455)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Core runs Claude Code through Paperclip Runner. Claude can start a
shell command in the background and end its turn. When the command
finishes, Claude Code starts a cycle of its own to continue.
> - Today the run ends when the turn ends. The command stops, and the
run fails with `native_finalization_missing`.
> - The Runner part of this work is complete in #21316 and the pull
requests below it: when a controller sends `providerStartedTurns` on
`turn.start`, the Runner keeps Claude while a background shell is
running, and gives the cycle of Claude to the controller as a turn that
the provider started.
> - Part 2 gave the consumer a wait with a limit and two inputs: a hook
that ends the wait early, and a start time.
> - This pull request gives both from the server for a native ACPX
Claude run. A new comment from a person ends the wait. The start time
comes from an event that is already saved, so a restart of Core does not
make the wait longer.
> - The benefit is that a person who comments does not wait up to 10
minutes for the agent to see it.

## Linked Issues or Issue Description

No public issue exists. The issue is described here.

This pull request is one of four parts on the Core side. They are
stacked on #21316:

1. The driver admits a turn that the provider started.
2. The consumer waits for that turn, for a limited time.
3. The server ends the wait when a person comments, and keeps the clock
across a restart.
4. The transport sends the opt-in. Only this part changes what a
production run does.

This is part 3. Its base branch is `feat/core-provider-turn-consumer`.
Refs #19912 (design).

**What existing behavior does this improve?**

Nothing visible yet. What Core gives to the consumer of a native ACPX
Claude run.

**Subsystem affected**

Server: `server/src/services/native-runtime`.

**Current behavior**

- A comment from a person during a run is queued as a deferred wake. The
next run delivers it. This does not change.
- Core has no wait for a provider-started turn.

**Proposed behavior**

- For a native ACPX Claude run, Core gives `executeNativeSession` the
option `providerTurnWait`.
- One delivery worker does the person-comment check for all runs that
wait. The server registers it at startup with the shared delivery work
coordinator, as it registers the cost accounting worker. The worker
listens to the existing notification for deferred wakes. On each notice
it runs one query for each run that waits: a comment from a person on
this issue, made after the start of the run, that a deferred wake of
this agent names. A comment that the query finds ends the wait of that
run. The run then takes the path of today, and the next wake delivers
the comment as today.
- If a database read fails, that run stays owed a check, and the
coordinator runs the worker again after 5 seconds, as for each other
delivery worker. A read that fails for one run does not stop the check
of the other runs. When no run is owed a check, the worker has no
deadline and does no query.
- The start time of the wait is the time of the first saved
`turn.completed` event of the run that has `awaitingProviderTurn`. With
no such event it is empty.
- Each query has the company id in its condition.

**Reason and benefit**

- The comment window starts at the start of the run, not at the start of
the wait. A comment made during the prompt turn must not wait for the
limit.

## What Changed

- `server/src/services/native-runtime/native-provider-turn-wait.ts`
(new, 109 lines) and 4 lines in `native-session-executor.ts`.
- `server/src/index.ts` (4 lines): the registration of the worker, its
stop at server close, and its place in the list of delivery workers that
must be ready. `delivery-work-notifications.ts` (1 line): the queue key
of the worker. The key needs no database change, because each topic uses
the one shared channel.
- `server/src/vendor/paperclip-runner/index.ts`: one type export.
- Tests: `native-provider-turn-wait.test.ts` (new, 252 lines) and one
case in `native-usage-accounting.test.ts` (25 lines).

## State saved by the previous release

- No new persisted state and no schema change. The start time is read
from `heartbeat_run_events`.
- **A run that the previous release started:** it has no waiting
terminal, so the start time is empty.
- **Core restarts during a wait:** the new process reads the start time,
so the limit is the original one. The consumer then continues the wait
(#21454), and the worker checks the comments again from the start of the
run, so a comment that was made before the restart still ends the wait.
- **Rollback of Core during a wait:** the previous Core stops at the
waiting terminal and finalizes the run as `native_finalization_missing`,
as today.

## Verification

At the head of this branch:

- `npx tsc --noEmit -p tsconfig.json`, `npx tsc --noEmit -p
tsconfig.test.json` and `pnpm run build:typescript` in
`packages/paperclip-runner`; `node
scripts/verify-runner-vendor-dependencies.mjs` and `pnpm typecheck` in
`server`; `node --test
scripts/__tests__/runner-release-crate-cache.test.mjs`: all passed.
- `native-provider-turn-wait.test.ts`, 10 tests on embedded Postgres
with the real coordinator: a comment from a person in a deferred wake
ends the wait; a comment from an agent, a comment before the run, and a
comment of a different company do not; the start time comes from the
first waiting terminal, and a run of the previous release has none.
- The 6 new tests: a read that fails one time is tried again after 5
seconds and finds a comment that is already saved, with no new notice; a
notice during a read that fails gives a new check; a read that fails for
one run does not hold back a second run, and each run ends only for its
own comment; a run that starts to wait during a check is checked; a wait
that is aborted is removed, nothing checks it again and no deadline
stays; with no run owed a check there is no deadline and no query in one
hour of test time.
- The test helper now registers the wake intent before it writes the
deferred wake, in the same transaction. This is the order of each
production writer (`modules/wake-queue/adapters/postgres.ts` and
`services/heartbeat/queue.ts`). Production code did not change for this.
- Mutation checks: a read that fails clears the owed check (3 tests
fail); an aborted wait is not removed (the abort test fails).
- Also run, unchanged and green: `delivery-work-coordinator.test.ts`,
`budget-enforcement-worker.test.ts`,
`server-startup-feedback-export.test.ts`, `startup-timeline.test.ts`.
The 7 files together: 1,699 passed, 7 skipped.
- `native-usage-accounting.test.ts`: a prompt turn and the
provider-started turn after it are each counted one time, with the
events as the executor receives them. The cost of the provider-started
turn is the session total in the cumulative channel, and it is not added
two times. The accounting code is not changed.

Not run: a run through the executor with a database; the full suite. CI
supplies the broad coverage.

## Risks

- **One more query at each start of a native ACPX Claude run,** on
`heartbeat_run_events` by company and run.
- **The worker is registered in each server process, and it is idle with
no wait.** A notice with no run that waits does no query. No wait starts
before part 4.
- **After an abort during a retry, the 5 second retry of the coordinator
runs one more time.** It finds nothing owed and does no query. The
coordinator has no call to cancel the retry of one worker.
- **A process that does not register the worker does no comment check.**
The server registers it with no condition at startup, and native runs
execute only in that process. Only a unit test can be in this state.
- **A comment ends the wait but does not interrupt a turn that is
running.** That is the rule of part 2.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`) in Claude Code planned the
change and the split, reviewed the diff, and ran a part of the
verification again.
- Claude Opus 5.5 (`claude-opus-5-5`), as a Claude Code subagent with
tool use (shell, file edits, local test runs), wrote the change and the
tests and cut this slice from the full change.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:55:11 -08:00
DottaandPaperclip 2102492806 fix(discord): refresh due command registration after task drains (#21573)
## Thinking Path

> - Paperclip runs agents and connects their work to chat services.
> - Discord command registration uses a shared worker and saved
deadlines.
> - A task drain can hold that worker while registration is already due.
> - The coordinator can keep a 30-second fallback after the drain ends.
> - This change refreshes the command owner on drain release or observed
expiry.
> - Normal admission, earliest-time and provider retry checks still
apply.

## Linked Issues or Issue Description

Refs: #17557, #21562. Related: #15532 checks credential reads when
registration is not due. This change handles resumption of due work in
the separate command worker.

**What happened?**

A due command registration can wait for the coordinator's fallback after
a task drain ends. The worker retains the request but has no
drain-release subscription. This delays command work and its return to
an idle state.

**Expected behavior**

Resume due command work when admission returns. Preserve future earliest
times and provider failure backoffs. Leave an empty or stopped worker
idle.

**Steps to reproduce**

1. Register the normal command worker while an idle task drain is
active.
2. Retain a due registration and capture the coordinator's fallback.
3. Release the drain, or observe its expiry, before that fallback.
4. Check that the due endpoint is checked before the fallback. Both
regression cases fail before this change.

**Paperclip version or commit**

Main base `3c15acd3fc1b200ef0e36ac904a03cd5fcba00c0`.

**Deployment mode**

Built from source. The focused tests use the real coordinator and
command owner with controlled registration operations. They do not
contact Discord.

## What Changed

- Subscribe the command owner to normal drain changes and observed
expiry.
- Wake it only when current admission allows work and its existing next
time is due.
- Remove both subscriptions through the normal worker stop callback.
- Add eight controls for release, expiry, another closed gate, future
earliest time, provider backoff, empty state and both stop paths.

## Verification

Local checks used the integrated branch that exposed this contract gap.
This main port contains the same two tested file afterimages. Its
admission, scheduler and fairness modules match those local inputs.
Main's coordinator and DB modules have other changes. Fresh CI on
`aec6ecc65717aa6372bad526e044ec4aa3888bec` now qualifies this main
context.

- Both release and expiry cases fail at the intended missing
endpoint-check assertion before the product change. No setup or timeout
failure is used as causal proof.
- All eight focused cases pass after the change. No case is excluded.
- The complete normal Server typecheck and affected-file program checks
pass on the integrated branch.
- Normal committed Source SDK verification passes on this main branch.
- Exact-head main CI passes, including typecheck, build, the general and
serialized suites, and all eight new controls (142 ms) in server shard
2/12. [CI
run](https://github.com/paperclipai/paperclip/actions/runs/43565579653).
A full local repository build and suite were not run in this checkout.
- Greptile reviewed the exact head at 5/5 with no actionable findings.
The complete edited summary and all paginated review threads were
inspected; no unresolved threads remain.

Run the focused controls with:

```sh
pnpm --filter @paperclipai/server exec vitest run src/services/chat-channels/discord/command-work.test.ts --maxWorkers=1 --no-file-parallelism --retry=0
```

The existing admission-drain contract remains accurate. This fix adds no
configuration or command, so no documentation change is required. Full
transport sleep acceptance and deployed suspension are separate claims.

## Risks

A release event can request an earlier pass for already due work. The
existing `canRun`, earliest-time, batching and per-item backoff checks
remain active. Both worker stop and coordinator stop must remove the
subscriptions; the controls cover each path. No scheduler policy, claim
or provider deadline changes.

## Model Used

OpenAI `gpt-6.1-sol` assisted with implementation and independent
review. The Codex managing agent performed technical review and
verification. The host does not expose its exact runtime model ID or
context-window size. Work used source inspection, code edits and local
test execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes (not
required: the existing contract remains accurate)
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-14 00:46:49 -06:00
DottaandClaude Opus 5.5 3c3e6c2d75 fix(retention): stop rescanning unchanged pinned workspace snapshots (#21571)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Remote workspaces keep saved versions in the
`execution_workspace_snapshots` ledger, and a retention sweep retires
versions that no reference holds
> - Every execution notification wakes this sweep, and the sweep
examined every pinned version again on each pass
> - Some pins never change. A migration cohort supersession leaves
failed operations with `lastFailure` and no `retiredSave`, and these
hold their snapshots until a separate review releases them
> - On a busy server the sweep ran four statements per pinned row, many
times each second, on a main thread that was already saturated
> - This pull request records the reference that pins a version and
leaves that version out of the candidate query until that reference
releases
> - The benefit is that an execution notification costs one candidate
query, not one transaction per pinned row

## Linked Issues or Issue Description

No public issue exists. Bug report:

**What happened?**
A hosted instance ran four statements at 13.4 calls per second each: the
workspace read, the importing-parent read and the operation read in
`pinned()`, and the locked ledger row read. The sweep in
`server/src/services/remote-workspace-snapshot-retention.ts` caused
them. A pinned superseded row keeps `status = 'superseded'` with
`delete_after = null`, so the candidate query selects it on every pass.
Each pass examines up to 50 rows. A truncated page runs again after one
second. Each run-work notification (`notifyExecutionRunWork`) sets
`rescanRequested`, which starts the scan again. The sequence stopped at
the operation read because a failed operation with `lastFailure` and no
`retiredSave` pinned these rows.
`supersedeExpiredRemoteWorkspaceMigrationCohort` writes such operations,
and they hold their snapshots permanently by design (see
`doc/DATABASE.md`, "Workspace cohort admission proofs").

**Expected behavior**
A pinned version is examined again only when the reference that pins it
can have changed. An idle or busy server does not repeat reads for
unchanged pinned history. A released version is retired on its normal
deadline.

**Steps to reproduce**
1. Create 300 superseded snapshots. Pin each with a failed
`remote_workspace_migration` operation that has `lastFailure` and no
`retiredSave`.
2. Register the retention worker on a delivery coordinator and let the
first pass finish.
3. Send `notifyExecutionRunWork` about 13 times each second for 60
seconds.
4. Count statements with `pg_stat_statements`.

**Paperclip version or commit**
`master` at 6dd0349d52

**Additional context**
The local reproduction showed the same four statements at equal call
counts, as on the hosted instance (see Verification).

## What Changed

- `pinned()` returns the first reference that holds the version (current
pointer, importing parent, operation, unfinished child, run, or recovery
action) and its ID, not a boolean.
- The sweep stores that reference in `metadata.retentionPin` when it
pins a version. It writes again only when the reference changes or a
deadline must be cleared.
- The candidate query leaves out a version with no deadline while its
recorded reference still holds. Each pin condition is one SQL builder.
`pinned()` and the candidate query use the same builder, so the
exclusion can never hold a version that `pinned()` would release.
- When the recorded reference releases, the version comes back, and
`pinned()` checks every reference again. It records the next reference
that holds, or schedules or retires the version.
- The record is removed when the sweep schedules or claims a version.
- The worker coalesces wakes for up to one second
(`REMOTE_WORKSPACE_SNAPSHOT_COALESCE_MS`). Retention deadlines are days,
so this delay has no effect on retirement.
- `doc/DATABASE.md` describes the record.
- No schema migration. The record is in the existing `metadata` JSONB
column.

## Verification

Red/green regression tests in
`server/src/services/remote-workspace-snapshot-retention.test.ts`:
- `leaves %s-pinned history out of later passes, across restarts, until
that pin is released`. This runs once for each of the six pin kinds. It
pins the version and checks that a new service instance (a restart)
examines zero rows. Then it releases the pin and checks that the version
is deleted.
- `re-examines a pinned snapshot when its recorded pin is released but
another still holds it`.
- `revisits a failed save's recorded pin once its abandoned upload is
settled`.
- `does not re-examine unchanged pinned history on execution
notifications`. This uses the coordinator and `notifyExecutionRunWork`.

On the parent commit, the first eight new tests fail (6 pin kinds, the
second-pin case, and the notification case). With the fix, all 31 tests
pass.

Related suites pass locally (212 tests):
`remote-workspaces.integration.test.ts`,
`remote-workspace-snapshot-reuse.test.ts`,
`remote-workspace-cohort-migration.test.ts`,
`snapshot-attempt-upload-abort.test.ts`, and
`remote-workspace-snapshot-retention.test.ts`. `tsc --noEmit -p
server/tsconfig.json` is clean.

Load measurement. Local PostgreSQL 18 with `pg_stat_statements`. 300
pinned rows. About 13 run-work notifications each second for 60 seconds,
after the first pass finished:

| | Before | After |
| --- | --- | --- |
| Pinned rows examined again | 30,250 – 39,000 / min | 0 / min |
| Sweep statements | ~151,000 – 196,000 / min | 88 / min |
| All statements, including the test's own notification transactions |
153,327 – 197,205 / min | 1,647 / min |
| Sweep database execution time | ~1,285 ms / min | ~52 ms / min |
| Candidate query | 0.07 ms, ~10 / min | 0.88 ms, 58 / min |

Before: `begin`, the locked row read, the workspace read, the
importing-parent read and the operation read each ran 30,250 times in 60
seconds. This is the same signature as on the hosted instance. After:
the sweep ran 58 candidate queries and 30 deadline reads. With
per-notification wakes and no coalescing, the sweep ran 776 candidate
queries and 776 deadline reads (1,552 / min).

I did not verify the change on a hosted instance.

## Risks

- A recorded pin is checked against only its own reference. The check
uses the same SQL condition as `pinned()`, so a version cannot stay
excluded after that reference releases. The parametrized test covers
each pin kind.
- A version can be pinned by more than one reference. When the recorded
reference releases, the sweep checks all of them again and records the
next one.
- A producer that rewrites the row's metadata and drops the record
causes one extra examination. A producer that sets a deadline makes the
record inactive, because the exclusion applies only when `delete_after`
is null.
- Versions that have no record yet, which includes all existing pinned
rows, are examined once after deploy. The examination writes the record.
- The candidate query does one primary-key lookup for each pinned row.
It takes about 0.9 ms with 300 pinned rows, and the time grows linearly
with the number of pinned rows.
- Pinned history from failed migration cohorts stays pinned. This change
does not alter that hold.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used through Claude
Code with tool use (shell, file editing, local PostgreSQL)

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 00:05:32 -06:00
Nicky LeachandClaude Opus 5.5 b1f608296e feat(runner): keep a run open for a limited time while the Runner waits for the provider's own turn (#21454)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Core runs Claude Code through Paperclip Runner. Claude can start a
shell command in the background and end its turn. When the command
finishes, Claude Code starts a cycle of its own to continue.
> - Today the run ends when the turn ends. The command stops, and the
run fails with `native_finalization_missing`.
> - The Runner part of this work is complete in #21316 and the pull
requests below it: when a controller sends `providerStartedTurns` on
`turn.start`, the Runner keeps Claude while a background shell is
running, and gives the cycle of Claude to the controller as a turn that
the provider started.
> - `consumeTurn` reads the events of one turn and returns at the
terminal. It already continues past a terminal for a Codex goal.
> - This pull request makes it continue past a terminal that waits for a
turn of the provider, and it limits that wait.
> - The benefit is that a run continues in the turn that Claude starts,
and a wait always ends.

## Linked Issues or Issue Description

No public issue exists. The issue is described here.

This pull request is one of four parts on the Core side. They are
stacked on #21316:

1. The driver admits a turn that the provider started.
2. The consumer waits for that turn, for a limited time.
3. The server ends the wait when a person comments, and keeps the clock
across a restart.
4. The transport sends the opt-in. Only this part changes what a
production run does.

This is part 2. Its base branch is `feat/core-provider-turn-driver`.
Refs #19912 (design).

**What existing behavior does this improve?**

Nothing in production yet. What the consumer of a native session does
after a turn that ended with no result while the Runner waits.

**Subsystem affected**

Paperclip Runner package, the controller side:
`src/native-session-runtime.ts` and a new module
`src/native-provider-turn-wait.ts`.

**Current behavior**

- `consumeTurn` returns at the first terminal. With no result, the run
is finalized as `native_finalization_missing`.

**Proposed behavior**

- At a terminal with `awaitingProviderTurn` and no result, `consumeTurn`
continues to read. It consumes the provider-started turn as the next
turn of the same run. A turn that reports a result finalizes as a normal
run.
- The wait has a limit of 10 minutes from the first terminal that waits.
A later terminal that waits does not start the clock again. The limit is
never later than 5 seconds before the timeout of the run.
- The limit and an early end stop the wait only. They never interrupt a
provider-started turn that is running; that turn stays under the timeout
of the run. If that turn ends and the Runner waits again, the run takes
the path of today at once.
- When the wait ends with no turn active, the consumer returns the last
terminal, and the existing code for a run with no result decides
(`resolveMissingResult`, then `native_finalization_missing`). No new
failure code.
- The wait ends at once when the driver gives the acknowledgement of
part 1. This covers Stop, the limit, and a Runner that says the wait is
over.
- `executeNativeSession` gets one new option, `providerTurnWait`: the
start time of a wait that an earlier controller process saw, and a hook
that ends the wait early.
- When the wait ends a turn, the consumer first waits for the receipts
of the text deltas that are committed in batches (#21460), as each other
return path of `consumeTurn` does. Without this, a delta that arrived
during the wait was not in the event count of the turn, and a delta that
failed to commit was not reported.
- When the interrupt that ends the wait is not available or fails, the
limit forces the end of the wait. If that forced end starts a new turn
(the continuation that a restart owes), the consumer continues to read
that turn. The forced-end signal is used one time only, and the read
that is open at that moment is kept until it gets its event. Before this
fix, the consumer returned the terminal of the old turn and lost the
first event of the new turn, and the run could end as
`native_finalization_missing`.
- If the controller restarts during the wait, the run continues to wait.
When the saved checkpoint still shows the first turn as active, the
recovered session gives the waiting terminal again, and the wait starts
again with the start time that was recorded. When the checkpoint was
saved after the waiting terminal, the consumer finds that terminal in
the journal, starts the wait again, and does not send the
disposition-only recovery prompt. The consumer does this only if two
conditions are true: the recovered session has the waiting state
(`awaitingProviderTurn`, saved by the driver since #21453), and the
terminal in the journal has the marker. If one condition is false, the
run uses the recovery path of today. When a wait that started again ends
with no turn, the existing code for a run with no result decides, as for
each other wait.

**Reason and benefit**

- The 10 minutes are the maximum time of a foreground command in the
pinned Claude Code.
- A limit on the wait, not on the work, means that a long test run in
the turn of Claude is not cut.

## What Changed

- `src/native-provider-turn-wait.ts` (new, 129 lines) and
`src/native-session-runtime.ts` (58 lines added, 4 removed),
`src/index.ts` (1 export).
- `src/native-session-runtime.provider-turn.test.ts`: written again for
the new behavior (557 lines added, 61 removed).

## State saved by the previous release

- This part adds no persisted state. The wait is in memory. Part 3 reads
its start time from an event row that is already saved.

## Verification

At the head of this branch:

- `npx tsc --noEmit -p tsconfig.json`, `npx tsc --noEmit -p
tsconfig.test.json` and `pnpm run build:typescript` in
`packages/paperclip-runner`; `node
scripts/verify-runner-vendor-dependencies.mjs` and `pnpm typecheck` in
`server`; `node --test
scripts/__tests__/runner-release-crate-cache.test.mjs`: all passed.
- `src/native-session-runtime.provider-turn.test.ts`: 30 tests. They
cover: a controller restart with the first turn still active in the
checkpoint (the wait continues; a provider-started turn settles its
result; with no turn the limit ends the wait on the recorded clock); a
controller restart with the checkpoint saved after the waiting terminal
(the same two results, no recovery prompt, no event of the journal
appended again); a limit that passed before the restart ends the wait at
once, with and without an interrupt; the recovery prompt of today when
the terminal in the journal has no marker, when the journal has no
terminal, and when the recovered session has no waiting state; a second
restart during the wait; a forced end that starts a new turn settles the
result of the new turn, with each event of that turn appended one time
and in order; an event that reaches the open read after the forced end
is appended exactly one time; a delta that arrives during the wait is
counted when the limit ends the turn (the test fails when the new wait
for receipts is removed); continue and settle the result of the
provider-started turn; a second wait does not start the clock again; the
limit with no turn active; the limit and a comment during a running turn
(not interrupted); a running turn that ends and waits again after the
limit; Stop; a `run.terminal` that the Runner held; the Runner stream
ends during the wait; the limit under the timeout of the run; a start
time from an earlier controller; a plan that a provider-started turn
proposes settles as the plan review wait.
- `src/native-session-runtime.test.ts`, not changed by this PR: 297
passed.
- Mutation checks, each made tests fail: no continue at a waiting
terminal; no limit; no early end; no clock from an earlier controller;
the clock starts again at each wait; a running turn is interrupted at
the limit.
- Mutation checks for the forced end, each made the two new tests fail:
the forced-end signal used on each loop (the run ends as
`native_finalization_missing`); a new read started after the forced end
(the first event of the new turn is lost).
- Mutation checks for the restart, each made named tests fail: the
recovery prompt not held back; the wait not started again in
`consumeTurn`; the wait started again on the session state alone (the
test with no marker in the journal fails).

Not run: the full suite. CI supplies that coverage.

## Risks

- **No production effect.** No terminal has the marker before part 4.
- **`consumeTurn` runs for each native turn of each agent.** It makes
one small object for each turn and reads one more payload field at a
terminal.
- **If the interrupt that ends the wait is not available or fails,** the
limit forces the end: the consumer returns the last terminal, or
continues to read when the forced end starts a new turn. This is a
fallback path. On a forced return, one read of the event stream stays
open until the session closes, as before this PR.
- **A wait that starts again after a restart holds back the recovery
prompt.** This is the one change to the recovery flow. It needs the two
conditions above, and both exist only for a run that asked for
provider-started turns.
- **Known limit.** The driver does not clear its waiting state when the
consumer ends a wait with an interrupt. A restart in the short time
after that can start the wait again; the recorded clock and the comment
check then end it.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`) in Claude Code planned the
change and the split, reviewed the diff, and ran a part of the
verification again.
- Claude Opus 5.5 (`claude-opus-5-5`), as a Claude Code subagent with
tool use (shell, file edits, local test runs), wrote the change and the
tests and cut this slice from the full change.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 05:59:42 +00:00
DottaandClaude Opus 5.5 3c15acd3fc perf(recovery): look up held runs by key so the fresh-continuation deadline stops scanning run history (#21568)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When a run is cut off, the fresh-continuation fallback continues the
task in a new run. The execution-disposition worker schedules its next
pass with `nextFreshContinuationCheckAt` and lists due work with
`heldExecutionsAwaitingFreshContinuation`
> - Both queries share two predicates (`heldExecutionWhere` and
`recentInterruptionSourceSql`). These predicates matched the run as
`source.id::text = <evidence text>` and the task's runs as
`native_issue_id::text = <issue id>::text`. A cast on an indexed column
prevents an index lookup
> - So for each held record, PostgreSQL read the company's whole run
history (two times), and it decoded each run's wide context JSON. On
staging, the deadline query took about 13.5 s of database time in each
27 s worker cycle. It was the largest statement by time
> - This pull request compares the uuid columns directly. It turns the
recorded run id into a uuid only when the text is the canonical
lowercase form, which is exactly when `id::text = text` can be true
> - The benefit is that each held record costs a few index lookups. The
same records are selected, and the worker cycle no longer keeps a
database session busy

## Linked Issues or Issue Description

**What existing behavior does this improve?**
The fresh-continuation scheduler queries: `nextFreshContinuationCheckAt`
(the disposition worker's deadline),
`heldExecutionsAwaitingFreshContinuation` (the backlog listing, also
probed by the schedule), and the "left alone" count. They are in
`server/src/services/fresh-continuation-fallback.ts`.

**Subsystem affected**
Server: recovery and the execution-disposition worker
(`executionControlSchedule().disposition` in
`execution-control-work.ts`).

**Current behavior**
Neon statement statistics on staging (window of 160 s on 2026-11-14):
the `select
min((issue_recovery_actions.evidence#>>'{freshContinuationCheck,nextCheckAt}')::bigint)
…` statement ran 6 times at about 13.5 s each. That is 81 s of database
time in 160 s, about half of one busy session. All other statements in
the window used less than 35 s together. Staging has 31,399 runs (1.96
GB, wide `context_snapshot`), 75,285 recovery records, and 282 records
with a check deadline.

The local plan shows the cause. For each candidate record:
- `source.id::text = coalesce(evidence->>'runId',
evidence->>'sourceRunId')` cannot use the primary key. PostgreSQL
filters the company's failed runs (6,767 rows removed in each loop, and
6,798 again in the outer semi-join).
- In the recent-interruption check, `max(held.finished_at)` matches runs
of the task with `native_issue_id::text = source_issue_id::text or
issue_id::text = … or context_snapshot->>'issueId' = …`. The casts
cannot use the task-link indexes, so PostgreSQL filters the company's
run history and decodes each run's context JSON (29,937 rows removed in
each loop).

**Proposed behavior**
The queries look up the source run by its primary key and the task's
runs by the existing `(company_id, native_issue_id, …)`, `(company_id,
issue_id, …)` and `(company_id, context_snapshot->>'issueId', …)`
indexes. The results do not change.

**Reason and benefit**
On a staging-shaped local fixture, the deadline query goes from 2,330 ms
to 27 ms (87x), and the backlog listing and count go from 3.7-5.7 s to
19-27 ms. Buffer accesses go from 2.15 M to 28 k.

**Breaking changes**
None. No schema change and no migration.

## What Changed

- `server/src/services/fresh-continuation-fallback.ts`:
- Add `sourceRunKey`. It compares the recorded run id with
`heartbeat_runs.id` through the existing `canonicalExecutionUuid` guard
(cast only when the text matches `^[0-9a-f]{8}-…-[0-9a-f]{12}$`).
`uuid::text` always gives that lowercase form, so this matches exactly
the rows `id::text = text` matched. Uppercase, braced, padded, numeric
and invalid ids still match nothing, and no cast error is possible.
- Use it in `heldExecutionWhere`, `interruptionEligibility` and the
listing's order expression.
- `runOfTask` compares `native_issue_id` and `issue_id` with the issue
id as a uuid. Both callers pass a uuid: the `source_issue_id` column, or
the id of a task row read in the same transaction.
- Export `heldExecutionWhere` and `recentInterruptionSourceSql` so the
regression test can compare them with the previous predicates row by
row.
- Add `server/src/services/fresh-continuation-schedule-queries.test.ts`
(regression test, see Verification).
- Add `scripts/benchmark-fresh-continuation-deadline.ts`. It seeds a
staging-shaped fixture, captures the SQL that the baseline revision and
the working tree send, compares complete positional rows, and reports
timings and plans.

## Verification

- Red/green: `fresh-continuation-schedule-queries.test.ts` "finds the
deadline by run-key and task-link lookups, not a scan of run history per
record" counts the `heartbeat_runs` rows that the deadline query's plan
reads (loops x (rows + rows removed)). With the previous predicates it
fails: 549,166 rows read against a bound of 9,442 (two times the run
count). With this change it passes.
- Equivalence: the same file seeds 900 held records with a seeded random
generator. The records cover these cases: `runId`, `sourceRunId`, both
keys, uppercase, braced, padded, numeric, invalid and unknown ids; a
source run in another company; source statuses (failed, timed_out,
interrupted, cancelled, succeeded, running) with and without
`finished_at`; the assignee and other agents; newer runs linked by
native id, legacy issue id, or context only; unrelated runs;
cancelled-at-dispatch runs; live (queued, running, scheduled_retry)
runs; earlier and later creation times; retries of the source; person
Retry wakes; old and recent finish times; restart, person-owned and
sandbox-resume error codes; and issue statuses and assignees. For every
record it compares `heldExecutionWhere()`, `heldExecutionWhere({
includeActive: true })` and `recentInterruptionSourceSql(now)` with
verbatim copies of the previous predicates. It also compares
`nextFreshContinuationCheckAt` and the full ordered
`heldExecutionsAwaitingFreshContinuation` rows with the previous
queries. The test checks that each predicate is true and false for more
than 5 records, so the comparison is meaningful.
- Focused suites pass locally: `fresh-continuation-fallback.test.ts` and
`fresh-continuation-schedule-queries.test.ts` (82 tests). Related
recovery suites pass: `fresh-continuation-retry-delivery`, `task-retry`,
`fresh-continuation-move-forward-offer`,
`execution-recovery-dead-controller`,
`automatic-run-brake-recovery-passes`, `execution-control-work`,
`recovery-wake-herd` (with the new test: 8 files, 252 tests). `node
server/scripts/typecheck-layers.mjs` passes.
- `pnpm verify` exits 2 (the new benchmark script is a global CI input),
so CI supplies the broad coverage.
- Benchmark: local embedded PostgreSQL 18.1, `jit=on`,
`shared_buffers=128MB`, macOS arm64, a shared machine with other work.
Baseline `67652580ee`. Fixture: 5,669 issues, 31,351 runs with 60 KiB
incompressible context JSON (1,962 MB), 75,285 recovery records (282
with a check deadline, a third overdue). One warm-up and 5 alternating
timed pairs. Both versions return identical positional rows for every
query.

| Query | Baseline median (samples, ms) | This PR median (samples, ms) |
Speedup | Shared buffers |
  | --- | --- | --- | --- | --- |
| `nextFreshContinuationCheckAt` | 2,330 (1,962-2,421) | 27 (21-28) |
87x | 2,149,802 -> 28,227 |
| `heldExecutionsAwaitingFreshContinuation` limit 5 | 3,692
(3,296-4,014) | 19 (18-24) | 194x | 2,831,539 -> 23,970 |
| same, limit 1 (schedule probe) | 3,554 (3,210-4,861) | 20 (18-24) |
179x | 2,831,539 -> 23,970 |
| left-alone count | 5,746 (5,090-5,992) | 27 (25-28) | 213x | 5,435,614
-> 40,953 |

With 1 KiB contexts (rows stored inline) the deadline query goes from
4,096 ms to 19 ms. Staging measured 13.5 s for the deadline query. The
local fixture reproduces the same plan shape, but the staging data
distribution and Neon storage are not reproduced exactly, so the staging
gain is not verified yet.

  ```sh
pnpm --filter @paperclipai/server exec tsx
../scripts/benchmark-fresh-continuation-deadline.ts \
--baseline-ref=67652580ee --context-kib=60 --samples=5 --warmups=1
--print-plans
  ```

- Plan after the change: the source run is an `Index Scan using
heartbeat_runs_pkey` (0 rows removed), and each task-run probe is a
`BitmapOr` over the native-id, issue-id and context-issue indexes. The
newer-run `not exists` already used that `BitmapOr` before. It does not
change.

## Risks

- Low risk. The SQL change is limited to how two keys are compared. The
regression test compares the old and new predicates for every record of
a varied fixture.
- `runOfTask` now casts its issue id to uuid. A caller that passes a
non-uuid text would get an error instead of no match. The two callers
pass a uuid column and a task row's id.
- Follow-up, not changed here: the disposition schedule calls
`deadline()`, which returns "retry in 15 s" when the earliest check is
already due. An active hold that the disposition pass does not pick up
(for example, one that records its run only as `sourceRunId`, which
`unresolvedExecutionDispositions` does not read) keeps the deadline
overdue, so the worker runs again every 15 s with nothing to do. A local
probe confirmed this. This explains a fixed cadence of 15 s plus the
query time (about 27 s on staging). After this change each such pass
costs milliseconds, not seconds.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used through Claude
Code with tool use (shell, file editing, local PostgreSQL benchmarks and
tests).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes (no
behavior or command change; no doc update needed)
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
canary/v2026.1114.0-canary.3
2026-11-13 23:37:19 -06:00
DottaandClaude Opus 5.5 3ae7aba643 perf(db): reuse dedicated connections instead of opening one per call (#21567)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The server uses `withDedicatedDbConnection` (packages/db) when an
action must hold one database session outside the main pool: advisory
transaction locks, lock keepalive, and `pg_settings` reads for workspace
lifecycle, finalization, source locks, skill caches and status cards
> - Each call built a new postgres.js client, opened a new physical
connection (TCP, TLS and startup through the pooler), built a new
drizzle client with the full schema, and then ended all of it
> - On staging the dedicated application name shows about 21 connection
setups per second, and the main thread is pinned (event-loop utilization
0.965)
> - This pull request keeps a small set of idle single-connection
clients per owner and reuses them
> - The benefit is that a dedicated action costs about 0.2 ms of
main-thread time instead of 4 to 8 ms, with the same exclusive-session
guarantee

## Linked Issues or Issue Description

**What existing behavior does this improve?**

`withDedicatedDbConnection` opens and closes one new database client and
one new physical connection for every call.

**Subsystem affected**

Database client (`packages/db`) and the server services that take
dedicated advisory-lock sessions.

**Current behavior**

Each call builds a new postgres.js client, opens a new connection (TCP,
TLS and startup through the pooler), and builds a new drizzle client
with the full schema. On staging (Neon statement statistics, 326 s
window) `SET application_name='paperclip-workspace-finalization-lock'`
ran 21.4 times per second. Only this factory uses that application name.
In a 30 s staging CPU profile, drizzle `relations.ts` was 2.2%
inclusive, `n:net` 5.8% and postgres.js construction 0.3%. Part of that
is this per-call setup.

**Proposed behavior**

A caller still gets exclusive use of one session for the whole action.
The session and its drizzle client are reused by later actions, so the
per-call cost is the action's own queries, not a full connection setup.

**Reason and benefit**

The server main thread is saturated on staging (event-loop utilization
0.965). Reuse removes about 5 ms of main-thread work per call.

**Breaking changes**

None. The exported function keeps its signature.
`closeDedicatedDbConnections` and `DEDICATED_CONNECTION_IDLE_MAX` are
new exports.

## What Changed

- `packages/db/src/client.ts`: each `createDb` owner now has one set of
dedicated single-connection clients, shared with every private client
derived from that owner (so nested dedicated actions use the same set).
- An action takes an idle client (last in, first out) or opens a new
one. It never waits. Callers hold advisory locks and nest dedicated
actions, so a waiting queue could deadlock.
- On success the client returns to the idle set if fewer than
`DEDICATED_CONNECTION_IDLE_MAX` (4) are idle. Otherwise it is closed.
- When the action throws, the client is closed and never reused. This
covers a failed or cancelled transaction, a broken socket, and an
interrupted lock wait.
- When an action ends the client itself (the source-lock cancel path
calls `$client.end`), the client is never reused.
- Idle clients keep the owner's idle timeout and max lifetime, so
postgres.js still closes idle physical connections after
`DATABASE_IDLE_TIMEOUT_SECONDS`.
- New `closeDedicatedDbConnections(db, timeoutSeconds)` closes the idle
set and stops retention. Ending the owner `$client` does the same. A
later action still works with a connection that closes on release.
- `server/src/index.ts`: shutdown closes the dedicated set next to the
other private pools.
- `doc/DATABASE.md`: documents the idle cap, the "not an admission
limit" rule, and the transaction-scoped session state contract.
- Session state audit: every current caller uses `pg_advisory_xact_lock`
/ `pg_try_advisory_xact_lock` and `set_config(..., true)`. No caller
uses session advisory locks, `SET` without `LOCAL`, `LISTEN`, or
temporary tables. A reset round trip per call is therefore not added.

## Verification

- New `packages/db/src/dedicated-connection-pool.test.ts` (embedded
PostgreSQL, 10 tests).
- Red on `origin/master`: 200 sequential calls used 200 distinct
sessions (`expected 200 to be 1`). All 10 tests failed.
- Green with this change, 3 runs in a row: 200 sequential calls use 1
session; 7 concurrent callers each hold a different session for the
whole transaction, and afterwards at most 4 stay open and later calls
reuse them; a failed transaction's connection is closed; a terminated
socket during the action is not reused; an idle connection terminated by
the server reconnects; a client the action ended is not reused;
transaction-local settings do not leak; shutdown and owner `$client.end`
close every retained connection; nested actions do not block.
- Focused suites pass: packages/db `work-signals`,
`prepared-queries-postgres`, `response-capture-postgres`,
`checkpoint-pool`, `client`, `client-options`,
`client-teardown-registry`, `postgres-connection-recovery`,
`client-disconnect-replay` (132 tests). Server suites that use dedicated
connections: `skill-package-store`, `status-card-evaluation-work`
(updated: it now proves advisory locks are released, since sessions are
reused), `agent-instruction-working-copies`,
`workspace-recovery-retry-copyback`, `native-execution-lease-batch`,
`company-skills-empty-runtime`, `decisions-service`,
`legacy-controller-lease`, `server-startup-feedback-export` (467 tests).
- `pnpm verify --explain` selects 2779 files for a
`packages/db/src/client.ts` change. Per AGENTS.md I ran the focused
checks above and let CI give the broad coverage.
- Benchmark: 200 calls of `withDedicatedDbConnection(db, d =>
d.transaction(tx => tx.execute(select pg_try_advisory_xact_lock(42))))`
against local embedded PostgreSQL 18, 10 warm-up calls, 5 alternating
before/after process pairs per mode. Main-thread time is
`eventLoopUtilization().active`.

| Mode | Variant | Wall ms/call (median, range) | Main-thread ms/call
(median, range) |
| --- | --- | --- | --- |
| TLS | before | 35.55 (32.93–37.52) | 5.18 (4.68–7.95) |
| TLS | after | 0.19 (0.17–0.21) | 0.18 (0.17–0.20) |
| Plain | before | 18.39 (17.91–20.53) | 3.65 (3.58–4.30) |
| Plain | after | 0.17 (0.13–0.18) | 0.13 (0.12–0.14) |

At the staging rate of about 21 calls per second, this removes roughly
100 ms of main-thread work per second (TLS figures; staging CPUs may be
slower).

## Risks

- A retained connection that the server or network drops while idle:
postgres.js reconnects on the next query when it has seen the close
(tested). If the drop and the reuse race, the first query of that action
fails like a main-pool query would. Before this change, a dedicated
action always started on a fresh connection.
- Through a transaction-mode pooler, `SET application_name=...` is a
server-side parameter switch, so its count may not drop as much as the
client-side connection count. Measure the deployed effect with pooler
client connections and main-thread CPU, not only that statement.
- Concurrency above 4 opens extra connections exactly as before; they
close on release. Total retained connections per owner grow by at most 4
idle clients, which still close after the idle timeout.
- An action that sets session-level state would now leak it to the next
action. The contract is documented on `withDedicatedDbConnection` and in
`doc/DATABASE.md`; no current caller does this.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used in Claude Code
with tool use (shell, file edits, test runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:37:04 -06:00
DottaandClaude Opus 5.5 c36b96ae00 fix(recovery): hold the start of queued automatic runs while a company's automatic-run brake is on (#21561)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The recovery system starts automatic runs: continuations, retries,
and sweep replacements
> - #21443 added a per-company automatic-run brake. It stops recovery
from creating new automatic runs while a company's runs keep failing
before they start
> - The brake did not hold automatic runs that were already queued. The
dispatcher still started them, so a burst of queued recovery runs kept
failing in the same way
> - This pull request adds the brake's start gate to the queued-run
dispatch pass, and wakes a held company at its next try without a
polling loop
> - The benefit is that a failing company stops spending runs on
automatic work, people's work still starts at once, and held work starts
again on its own when the brake releases

## Linked Issues or Issue Description

Refs #21443
Refs #21433
Refs #21427
Refs #21505

**What happened?**
While the automatic-run brake was open for a company, its already-queued
automatic runs (continuations and retries) still started. Each one
failed before it started, for the same cause that opened the brake.

**Expected behavior**
A queued automatic run must stay queued while the brake is open. Only
the brake's single probe, and then the half-open release rate (2, 4, 8
per window), can start one. A person's work (a Retry, or a continuation
of a person's request) must never be held. Held runs must start again on
their own, with nobody acting.

**Steps to reproduce**
1. Make 20 automatic runs in one company fail before they start, so the
brake opens.
2. Queue automatic continuations for that company.
3. Run a dispatch pass. Before this change, the continuations start.

## What Changed

- `server/src/services/heartbeat/queue.ts`: the dispatch pass asks
`reserveAutomaticRunStart` after the tried/backoff skip and the
claim-attempt budget check, and before the budget increment and the full
row read. A held run spends no claim attempt, does no row read, and does
not enter the declined-claim backoff. If the brake grants the start (its
probe or a half-open release start) but the claim then returns null,
throws, or finds the row gone, the pass gives the start back, so another
automatic run can use it in the same pass. The dispatcher's
person-origin answer is passed as is, and it is `undefined` only when
the dispatcher did not check origin (an agent with one queued run). Then
the brake's own cached person check decides.
- `server/src/services/heartbeat/queued-run-line.ts`:
`queuedRunDispatchFacts` also returns `personOriginChecked`, so the
queue knows when the dispatcher's negative answer is real.
- `server/src/services/heartbeat/dispatch-work.ts`: the queued-run
dispatch deadline worker adds companies whose next try is due to its
scopes, and includes the next try time in its next wake. A held company
wakes for its next try without a polling loop. If the brake's bounded
wake list had to drop a company, the worker wakes all companies at the
earliest dropped time.
- `server/src/services/recovery/automatic-run-brake.ts`: when the brake
holds a run, it records the company's next try time (probe time or the
next release window) in a bounded map. A probe or half-open release
grant returns a release that gives the start back with a compare-and-set
on its own run or window. It exports `reserveAutomaticRunStart`,
`takeDueAutomaticRunWakes`, and `nextAutomaticRunWakeAt`. The brake now
uses `isRecoveryRun`, `isPersonOwnedFailureCode`, and
`isRestartClassFailureCode` from `continuation-cause.ts` directly. The
rules are the same as before; the lazy loader keeps only the imports
that would make an import cycle.
- `server/src/__tests__/automatic-run-start-gate.test.ts` (new, 11 tests
through the real heartbeat queue): bounded queries for 250 held runs
over 100 passes, and for 1,500 held runs (past the 1,024-run cache) over
3 passes; a declined probe claim gives the probe back and another
automatic run starts in the same pass, and the probe stays with that run
on a later pass; an unused half-open start goes back to its window; with
276 held companies (past the 256-company wake list) every company is
still woken; a work run behind 250 held runs starts in the same pass; a
person's Retry is never held; the only queued run is not held when it
continues a person's Retry; exactly one probe starts across two server
instances; the release gives 2, 6, 14 cumulative starts over three
windows; held runs are not in the declined-claim backoff after the
pause; the dispatch worker wakes the company at the next try; and an
end-to-end test that a held automatic run stays queued while the brake
is open, then starts and succeeds on its own at the next try.
- Two small test updates in `automatic-run-brake.test.ts` and
`automatic-run-brake-import.test.ts` for the new cache size field and
the moved rule.

- `server/src/__tests__/automatic-run-brake-recovery-passes.test.ts`:
one test for the sweep path from #21505 (continue a blocked task behind
a run refused at dispatch). The sweep only records the continuation (no
wake, no new run), and `deliverReconciledExecutions`, which the brake
already gates, creates the run only after the pause. No second gate is
needed on that path.

## Verification

- `npx vitest run src/__tests__/automatic-run-start-gate.test.ts`: 11
passed.
- Red before for the review fixes: on the first PR head, the 1,500-run
cost test made 3,848 queries (bound 130), the declined-probe test
started 0 runs (expected 1), and the 276-company wake test missed 20
companies.
- Red before: with the `queue.ts` gate line removed, 5 of the 8 tests
fail (bounded pass, single probe, release rate, backoff, end-to-end).
- Red before for the person case: with the dispatcher's `false` passed
through as final, "never holds the only queued run when it continues a
person's Retry" fails (0 started). It passes after the change.
- `automatic-run-brake.test.ts` 19 passed,
`automatic-run-brake-import.test.ts` 1 passed,
`automatic-run-brake-recovery-passes.test.ts` 28 passed,
`failed-continuation-continues-again.test.ts` 32 passed (after merging
master with #21505).
- Nearby queue tests (on the final head): `heartbeat-run-dispatch-work`
15, `queued-run-dispatch-stacking` 6,
`heartbeat-queued-run-claim-isolation` 10,
`assigned-work-stalled-queued-run` 12, all passed.
- `server` `tsc --noEmit` is clean.

## Risks

- Behavior change only while a company's brake is open or half-open.
When the brake is closed, the gate returns from its cache with no query.
- The first time the gate sees a non-person automatic run while the
brake is on, the brake does one person-origin query for that run. The
result is cached per run (bounded to 1,024 runs).
- No migration. No API or UI change.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, in Claude Code with
tool use (shell, file edits, test runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:33:53 -06:00
DottaandClaude Opus 5.5 84cb28f9b7 fix(connections): keep machine sign-in secrets and settings in one locked commit (#21566)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - #21473 added machine sign-in (MCP OAuth client credentials) for
organization connections
> - Auto-merge squashed #21473 from an earlier head, so its last review
fixes did not reach master
> - Those fixes close a P1 race and two smaller setup gaps that Greptile
found on #21473
> - This pull request applies that last commit to current master, with
its tests
> - The benefit is that a client ID and its secret always stay together,
and key switching and app selection work as intended

## Linked Issues or Issue Description

Refs #21473

**What existing behavior does this improve?**
Machine sign-in setup and rotation, added in #21473.

**Problem**
- A connection test wrote a replacement client secret before checking
whether the connection changed. If the save was then refused, the secret
stayed and could pair with another client ID, so the next token request
failed.
- Switching a client-secret connection to a Paperclip key always created
an RS256 key, even when the sign-in service accepts only ES256.
- Machine sign-in dropped the setup's app ID and created a separate app.

**Proposed improvement**
Write the secret and the settings in one locked transaction after the
change check. Record the accepted signing algorithms and use them. Pass
the app ID through.

## What Changed

- `client-credentials-service.ts`: `saveConnection` locks the row,
checks that the sign-in method and credential references are unchanged
and that the connection is not removed, then writes any new credential
and the settings in the same transaction. Test, secret rotation and key
staging all use it.
- Setup records `tokenEndpointAuthSigningAlgValuesSupported`. Switching
from a secret to a key uses `chooseAssertionAlgorithm` on it.
- UI: `MachineSignInSetup` takes `applicationId`, and the custom-server
flow passes `prefill.applicationId`.
- Tests: two new server tests and one UI expectation.

## Verification

- `server/src/__tests__/tool-access-client-credentials.test.ts`: 25/25
pass. The 2 new tests fail on master's service code ("never leaves a
refused test's secret behind with another client ID", "creates the kind
of key the sign-in service accepts when switching from a secret").
- `MachineSignInSetup.test.tsx` and
`GenericMcpConnectStep.machine.test.tsx`: 8/8. The app ID expectation
fails on master's component.
- Server `tsc --noEmit` clean. UI `tsc` clean except the pre-existing
generated Storybook catalog imports.

## Risks

Low. The change is limited to machine sign-in. A concurrent change now
returns 409 `client_credentials_changed` instead of saving partly. No
migration.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, test runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 05:30:14 +00:00
DottaandClaude Opus 5.5 bbed32450c perf(load-model): model recorded native run rates, the recorded event mix and coalesced delta batches (#21549)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - `scripts/benchmark-agent-scaling-load.ts` (#21418) estimates how
many concurrent agents one server process can carry
> - It assumed 20 text deltas/s and 1 other event/s per agent, and every
other event was an `item.completed`
> - Recorded hosted native runs show about 0.2 deltas/s and 0.5 other
events/s, with a mix in which 39% of the other events are server trace
spans that take a cheaper path
> - This pull request makes the recorded rates and mix the model's
default, keeps the old rates as an explicit stress profile, adds a burst
profile, and mirrors the delta batching of #21520
> - The benefit is a model that measures what production does, so the
next optimizations target the real per-agent cost

## Linked Issues or Issue Description

Refs #21418 (load model and plan). Refs #21520 (delta batching that the
model now mirrors).

**What existing behavior does this improve?**
The load model's default rates, its event mix and its delta batching
rule, and the plan document's measurements.

**Subsystem affected**
`scripts/benchmark-agent-scaling-load.ts`, its test
`server/src/__tests__/agent-scaling-load-model.test.ts`, and
`doc/plans/2026-11-13-thousand-concurrent-agents.md`.

**Current behavior**
The model defaults to 20 deltas/s and 1 other event/s per agent. Every
other event is a native `item.completed` with a chat-progress wake and a
checkpoint save. It cuts a delta batch on the timer even while the
previous batch commits.

**Proposed behavior**
- `PROFILE=recorded` (default): 0.2 deltas/s and 0.5 other events/s per
agent. `PROFILE=burst`: 2.5 and 3, a run's busiest minute.
`PROFILE=stress`: the earlier 20 and 1. API requests stay at 0.3/s. Each
rate can still be set on its own.
- `EVENT_MIX=recorded` (default) sends other events in the recorded
proportions. A `run.performance.span` is an ordinary append with no wake
and no checkpoint, as the server's run trace writes it.
`tool.execution.*` and `item.completed` are native appends with a
chat-progress wake, then a checkpoint save. `EVENT_MIX=<type>` sends one
type only.
- Delta batching follows #21520: deltas that arrive while a batch
commits all join one next batch.
- The report adds `profile` and `eventsIssuedByType`.

**Reason and benefit**
At the recorded rates the per-agent cost is the non-delta runner events
and their checkpoint saves, not deltas. The plan and the next
optimizations need that picture.

**Breaking changes**
None for the product. Anyone who runs the model with no options now gets
the recorded profile; `PROFILE=stress` gives the earlier load.

## What Changed

- Profiles, the recorded event mix (a 100-slot cycle per agent, offset
by agent so the fleet sends the mix at every moment), the span append
path, and the coalesced delta batches.
- The model test now checks the profile in the report and that both
server trace spans and runner events with a progress wake are in the
mix, for the recorded profile and for the stress profile with one
transaction per delta.
- The plan has a new section "Load model at recorded rates": the
recorded source (40 recent native runs on one hosted workspace, as
aggregates), the profiles, the mix and its server paths, fresh runs at
500, 1,000 and 2,000 agents plus burst and stress, and what they show.
It records the 3-connection checkpoint pool as a computed per-process
limit on a hosted database, and that SQL building in Drizzle costs about
as much main-thread time as result parsing on the hosted instance. A new
section "Considered and not done" records why `run.performance.span`
rows stay (about 3% of the statements per agent, with readers) and why
checkpoint saves per critical event only are not done now.

## Verification

- `server/src/__tests__/agent-scaling-load-model.test.ts` passes (2
tests). The new assertions fail against the previous script: it reports
no `profile` and no `eventsIssuedByType`.
- `server/src/__tests__/agent-scaling-load-model.test.ts` also checks
each profile's rates with no overrides, and that a rate set on its own
wins.
- Load model runs, recorded in the plan (60 s, main pool of 10, embedded
PostgreSQL on the same machine, each run started below a 1-minute load
average of 15): recorded rates at 1,000 agents (event-loop utilization
0.61) and 2,000 (0.85) finish every operation with no failure; 500
agents all in their busiest minute run at 0.95 with p99 lateness under
0.4 s. About 4.4 statements per agent per second at recorded rates, 35%
of them on the checkpoint pool.
- Earlier runs at load averages of 20–55 had shown one process falling
behind at 500 burst agents and at 2,000 recorded; the quieter runs after
the model fixes (event order, spans off the runner stream) do not, and
the plan says so. The checkpoint pool as a per-process limit is now
stated as a computed limit for a hosted round trip (about 400 agents in
a busy minute at 1 ms), not as measured.
- Greptile findings fixed: a waiting delta batch no longer takes deltas
sent after the next runner event, takes every delta queued when it
starts, and spans no longer join the runner event stream.

## Risks

- Low: a benchmark script, its test and a plan document. No product code
changes.
- The machine that measured the plan's table was shared; the plan marks
the rows that the machine distorted.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:13:07 -06:00
DottaandClaude Opus 5.5 6dd0349d52 fix(ui): name a queued turn's hold once, in the turn header (#21564)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - A task's chat shows its run as a turn: a header line with the run's
state, then the turn body
> - #21405 and #21447 made a queued turn name why it waits: "Waiting for
Codie · …", "Waiting for PAP-5656 to finish", "Waiting to start"
> - On staging, the turn body repeated that label under the header
("Waiting for PAP-5656 to finish" twice). The tail used by other
adapters showed a fixed "Waiting to start..." even while the pill above
said "Working" or named a hold
> - #21447 removed the repeat only for the agent's waiting line
> - This pull request names every queued hold once, in the turn header
> - The benefit is one clear statement of why a task waits, with no
repeat and no contradicting line

## Linked Issues or Issue Description

Refs #21405, #21447

**What happened?**

On staging (revision 6b631dd47), a routine task waiting on the routine's
open execution showed "Waiting for PAP-5656 to finish · 31s" in the turn
header and "Waiting for PAP-5656 to finish" again in the turn body.
Other server holds (maintenance, agent setup, server restart, previous
run) and "Waiting to start" had the same repeat. For adapters other than
the Paperclip runner, the tail under the live run pill always said
"Waiting to start..." for a queued run, even while the pill said
"Working" (the 30-second startup feedback) or named a hold.

**Expected behavior**

A queued turn names its hold once, in the header. The turn body adds no
second line for it.

**Why does it matter?**

A repeated or contradicting line makes the waiting state harder to read.
DESIGN.md asks for one name per state.

## What Changed

- `TaskChatRunnerTurn`: a queued turn that has a run and a hold label
(agent queue, server hold, or "Waiting to start") no longer renders the
current-activity line under the header. Startup preparation before a run
exists ("Taking longer than usual…") keeps its slot, as DESIGN.md
requires.
- `TaskChatThread`: the live tail for other adapters no longer shows the
fixed "Waiting to start..." for a queued run. The live run pill above
names the wait.
- DESIGN.md: a queued turn's hold is named once, in the turn header.
- Storybook: the `WaitingForRoutine` story renders the full turn and
checks that the hold appears once.

## Verification

- Red/green: `TaskChatRunnerTurn.test.tsx`. The server-hold cases
(maintenance, agent setup, setup failed, server restart, previous run,
and the new routine hold) and the "Waiting to start" case now assert
that the label appears exactly once and that the turn body has no
current-activity line. Without the change, the 6 existing cases fail.
With it, all 52 pass. The preparation-label case still passes (its slot
stays).
- `vitest run` on `src/components/task-chat` and
`TaskChatThread.test.tsx`: 1,436 passed. The UI typechecks.
- The updated stories render in headless Chromium; the routine story
shows the hold once.
- After deploy: check on staging.

## Risks

- Low risk. This is a UI-only change. A queued turn with a hold shows
one line fewer. Running turns, terminal turns and pre-run preparation
are unchanged.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with tool use
(shell, file edits, headless browser).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 23:07:08 -06:00
DottaandClaude Opus 5.5 351b69941c perf(checkpoint): write a routine controller checkpoint save in one statement (#21553)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - A remote runner's controller saves a checkpoint after each runner
event it commits, on a 3-connection pool, before it acknowledges the
event
> - Each save is a transaction of 4 statements (5 where commit
notifications are on): BEGIN, the owner lock with the transaction id,
the write, COMMIT
> - What forces the transaction is the live-run recovery wake that every
save registers. Recovery does not react to a routine save
> - This pull request writes a routine save (new or changed event,
command and cursor records) with one statement that checks the owner and
writes, and keeps today's transaction for every other save
> - The benefit is one statement instead of four for most checkpoint
saves, on the pool that limits one process in a busy minute

## Linked Issues or Issue Description

Refs #21528 (owner lock returns the transaction id). Refs #21549 (load
model at recorded rates; the plan records the checkpoint pool as a
per-process limit).

**What existing behavior does this improve?**
`nativeRunnerCheckpointWriter` in
`server/src/services/native-runtime/native-runner-checkpoint.ts`.

**Subsystem affected**
Controller checkpoint persistence for remote native runs.

**Current behavior**
Every save locks the coordinator row (`native_run_finalizations` `for
update`), checks the owner, calls `notifyLiveRunRecovery` (6 recovery
queues), and upserts the changed parts, in one transaction.

**Proposed behavior**
A routine save is one statement: a CTE locks the coordinator row only
while the lease owner, boot id, generation and an unexpired lease
(`clock_timestamp()`) all match, and the upsert writes only through that
row. It registers no recovery wake. If the owner does not match, it
writes nothing and the save takes today's full path, which runs the
complete owner check (successor rules included) and fails with
`native_runner_checkpoint_owner_lost` when it should.

A save is routine when it has changed parts, no reset, no removed parts,
no header change, no predecessor, and the run is its own authority.
Every other save (the first save of a controller, which is a reset, a
save after a failed save, header or retirement changes, removals,
source-run consumers) is unchanged.

**Reason and benefit**
At recorded rates about a third of one process's statements are
checkpoint saves. In a busy minute the 3-connection pool cannot keep up
at 500 agents (#21549).

**Breaking changes**
None. No schema change.

## What Changed

- `ROUTINE_CHECKPOINT_SAVE_SQL` and the routine branch in
`nativeRunnerCheckpointWriter`.

## Why a routine save needs no recovery wake

I read the six consumers of `notifyLiveRunRecovery`:
- `nativeProcessSessions`, `nativeQuestionCancellations`,
`activeExecutionRecovery`, `settledNativeProcesses`
(`live-recovery-work.ts`): each selects candidates from `heartbeat_runs`
status, runtime mode, `native_run_finalizations` phase, lease and retry
times, the context snapshot and process identity. None reads checkpoint
rows to decide eligibility.
- `strandedAgentStatus` (same file, `agents` rows) and
`activeRunSilence` (`active-run-watchdog/adapters/deadline-work.ts`,
`heartbeat_runs.last_output_at`): neither reads checkpoint rows.
- The checkpoint rows matter to recovery in two places:
`listExpiredNativeRunnerAuthorities` requires that a checkpoint exists,
and the remote bootstrap proof reads its parts. A checkpoint first
appears with a reset save, which keeps its wake. A routine save only
adds event and command parts, which can only take a run further from the
bootstrap proof; recovery reads the current rows when its own deadline
(lease expiry) fires.
- The intent of the wake also fenced a recovery pass against a save in
flight. The routine save is one atomic statement and holds the
coordinator row lock while it writes, so a recovery claim that takes
that row waits for it, and a reader sees the save whole or not at all.

## Verification

- Red/green on embedded PostgreSQL
(`native-runner-checkpoint-save.test.ts`); the new tests failed against
`origin/master` first:
- A routine save is one statement (`with owner as ... for update ...
insert into native_runner_checkpoints`) and registers no recovery wake
(before: 4 statements and a wake).
- A reset save keeps BEGIN, the owner lock with the transaction id, and
COMMIT, and still registers its recovery wake.
- A routine save that waits while another connection takes over the
lease (new owner and generation) falls back to the full path and fails
with `native_runner_checkpoint_owner_lost`; the checkpoint is unchanged.
- The same race where the new owner also resets the checkpoint (deletes
every part and writes its own header): the save fails the same way and
the checkpoint holds only the new owner's header.
- A lease that expires while the routine save waits for the coordinator
row lock (the holder does not change the row) writes nothing: the
statement judges the lease after the lock, as
`lockNativeRunnerCheckpointOwner` does. My first version checked the
lease in the locking CTE itself, and PostgreSQL does not re-check a
condition after a lock wait on an unchanged row; the existing
`native-runner-restart-recovery.integration.test.ts` test "fences
checkpoint delivery when ownership expires during an unchanged row-lock
wait" failed on it, and so does the new test. Both pass now.
- The existing test "refuses a save after the owner's lease expired and
writes nothing" passes through the new fallback.
- Mutation checks: with the owner conditions removed from the statement,
both race tests and the expired-lease test fail; with the lease check
moved back into the locking CTE, the lock-wait expiry test fails.
- Other focused suites, one file at a time:
`native-runner-restart-recovery.integration` (231 passed, 14 skipped),
`native-runner-restart-checkpoint-cost.integration` (1; it greps for
`insert into "native_runner_checkpoints"`, so the statement quotes the
table name), `native-recovered-owner-followup` (97),
`native-controller-recovery.integration` (30),
`native-runner-checkpoint-authority.integration` (1),
`native-restart-recovery-isolation` (1), the port suite with conformance
(105), `heartbeat-native-restart-suspension` (115), `live-recovery-work`
(51 tests pass; on this busy machine its teardown hook timed out, as it
has before on master). `pnpm --filter @paperclipai/server run typecheck`
passes.
- Statements per routine save: 4 → 1 (5 → 1 where commit notifications
are on). I did not rerun the load model for this change: the machine is
too busy for a fair timing comparison today, and the statement count is
exact.

- CPU per operation (a local script, not committed: 600 routine saves of
one event part and the cursor for one run on embedded PostgreSQL, one
connection, three rounds, `origin/master` code against this branch):
main-thread CPU per save 0.38–0.56 ms before and 0.10–0.12 ms after
(encryption included in both); wall time per save 0.45–0.73 ms before
and 0.22–0.30 ms after.

## Risks

- A routine save no longer wakes the six recovery queues. Their own
deadlines and the 30 s reconstruction still run; see the reading above
for why no decision depends on a routine save.
- The owner check of the routine statement accepts only the exact active
owner. Any other state goes to the unchanged full path, so a successor
or a finished result keeps today's rules.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:58:06 -06:00
DottaandClaude Opus 5.5 67652580ee perf(tool-access): index enabled policies so each decision checks only candidates (#21563)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Tool access policies decide whether an agent can call each tool, and
an MCP tools/list response decides access once for each listed tool
> - Each decision matched the raw selector of every enabled policy two
times, so one listing cost tools x policies selector checks on the
server's main thread
> - On staging, a company with 363 enabled policies kept the main thread
at 95-98% of a core: policy matching was 22% of CPU time, with single 80
ms runs inside the selector match
> - This pull request compiles the enabled policies once for each
decision cache and indexes them by the selector key that rejects most
(catalog entry, then tool name)
> - The benefit is that each decision checks only the policies that can
match its tool, with the same decisions, so a listing no longer blocks
the event loop

## Linked Issues or Issue Description

**What existing behavior does this improve?**
Tool access decisions for MCP tool listings
(`toolAccessPolicyService.decide`, called by `decideToolsForListing` in
the tool gateway).

**Subsystem affected**
Server: tool access policy evaluation
(`server/src/services/tool-access-policy.ts`).

**Current behavior**
Each decision loops over all enabled policies two times. The first loop
matches the selector and schema-checks the conditions of every
conditioned policy. The second loop evaluates conditions and matches the
selector of every policy. The selector match reads and allocates every
selector key in a fixed order, so a catalog entry mismatch is found only
at the end. A listing of 800 tools with 365 policies does 305,600
selector evaluations. On staging this was 22% of main-thread time
(event-loop utilization 0.965, delay p99 144 ms, max 710 ms).

**Proposed behavior**
The decision cache holds the enabled policies compiled once: parsed
selectors, an index by catalog entry and tool name, and a memoized
conditions schema check. Each decision evaluates only the candidate
policies, in the original priority and creation order.

**Reason and benefit**
Listing cost becomes proportional to the policies that can match each
tool. In the benchmark, one listing goes from about 164 ms to 12 ms of
main-thread time, and the longest event-loop block goes from 84 ms to 4
ms.

**Breaking changes**
None. The policy data model, API, and listing contract do not change.

## What Changed

- Add `server/src/services/tool-access/policy-selector-index.ts`. It
compiles a policy selector into a matcher with the same semantics as the
raw `selectorMatches`, and it indexes policies by `catalogEntryId`, then
by `toolName` (matched against the tool name and the upstream tool name)
for policies without a catalog entry selector. Policies with neither key
are candidates for every call. Candidates keep their original order.
- `decideInternal` caches the compiled index under the same
`enabled-policies:${companyId}` key. The new `matchEnabledToolPolicies`
keeps the two-step semantics: the first selected policy with unsupported
conditions denies the call before any match applies, then the selected
policies with matching conditions go to the unchanged decision loop
(block, rate limit, trust rule, require approval, allow, explanations).
- Decisions without a cache, and decisions that consume a rate limit
(which read fresh rows), compile the rows they read in the same way.
- The conditions schema check runs once per policy and cache, not once
per decision.
- Conditions are now evaluated only for policies whose selector matches.
Before, a malformed legacy condition on an unrelated policy could throw
in every decision. A selected policy with such conditions still gets the
unsupported-conditions deny first.
- Add `server/scripts/benchmark-tool-policy-matching.ts` and a section
in `doc/mcp-discovery-performance.md`.

## Verification

- Red/green regression:
`server/src/__tests__/tool-access-policy-index.test.ts` "reads policy
selectors once per decision batch, however many tools it decides". It
decides 40 tools with one cache against a staging-shaped set of 365
policies and counts reads of the raw policy selectors. On
`origin/master` it fails: 308,110 reads for 40 tools against 7,711 for
the first tool. With this change the count stays at the first-tool
count. The test also checks that cached and uncached decisions are
equal.
- Equivalence: the same file compares the compiled selector with the raw
`selectorMatches` for 404 selectors x 300 contexts (including empty,
whitespace, non-string, string-plural, duplicate and contradictory
values). It compares the indexed policy selection with the old unindexed
two-step selection (copied into the test) for 5 random sets of 400
policies x 400 contexts, and for the staging-shaped set over 800 tools.
The sets include catalog-entry-only, catalog-entry+connection,
multi-key, tool-name, unindexed, conditioned, unsupported-condition,
block, rate_limit, trust_rule, allow and legacy policy types with mixed
priorities.
- Index edge cases: an earlier tool-name policy (matched through the
upstream name), an unindexed policy and an empty catalog-entry selector
apply before a catalog-entry block; an unsupported-condition policy in
another bucket still denies. A DB-backed test checks the same through
`decide`, with and without a cache.
- Focused suites pass locally: `tool-access-policy-index`,
`tool-access-policy-service`, `tool-gateway-listing-memory`,
`connector-tool-policy`, `tool-gateway`, `tool-gateway-service` (6
files, 406 tests). `node server/scripts/typecheck-layers.mjs` passes.
- `pnpm verify` exits 2 (the new benchmark script is a broad input), so
CI supplies the broad coverage.
- Benchmark, local embedded PostgreSQL, macOS arm64, Node 24. One
listing of 800 tools, 365 enabled policies (staging shape plus the 2
fixture policies), 16 concurrent decisions with one cache as in the
gateway. Three runs of 9 alternating rounds after 2 warm-up rounds. Run
3 uses the event-loop delay histogram that also covers the end of the
listing. Both versions return identical decisions.

| Per listing | `origin/master` median (min-max) | This PR median
(min-max) |
  | --- | --- | --- |
  | Elapsed, run 1 | 157.5 ms (146.1-192.8) | 11.4 ms (10.3-15.3) |
  | Elapsed, run 2 | 165.3 ms (161.7-185.4) | 11.9 ms (10.1-14.0) |
  | Process CPU, run 1 | 160.5 ms (149.7-188.9) | 13.7 ms (11.1-21.0) |
  | Process CPU, run 2 | 162.8 ms (157.2-192.7) | 12.3 ms (9.6-20.5) |
  | Elapsed, run 3 | 163.9 ms (151.6-185.3) | 12.4 ms (10.4-18.5) |
  | Process CPU, run 3 | 165.4 ms (155.7-214.9) | 14.5 ms (10.6-20.2) |
| Longest event-loop block, run 3 | 84.1 ms (76.9-95.1) | 3.7 ms
(3.0-5.0) |
  | Selector evaluations | 305,600 | 1,164 |
  | Raw selector property reads | 6,155,830 | 9,490 (compile only) |

  ```sh
cd server && pnpm exec tsx scripts/benchmark-tool-policy-matching.ts
--baseline-ref origin/master --tools 800 --rounds 9
  ```

- Not verified: the hosted staging profile after deploy.

## Risks

- Low risk. The decision loop is unchanged; only the set of policies
that reach it is narrowed, and every left-out policy has a selector that
rejects the call. The equivalence tests compare against the old
selection.
- The index lives only in the per-listing decision cache (or for one
uncached decision). There is no cross-request cache, so policy edits
apply on the next listing as before.
- The cached index memoizes the conditions schema check on shared
entries. The check is pure, so concurrent decisions in one listing get
the same result.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, local test and benchmark runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:51:43 -06:00
Nicky LeachandClaude Fable 5.2 82ce8e64ba feat(runner): let the session driver admit a turn that the provider started (#21453)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Core runs Claude Code through Paperclip Runner. Claude can start a
shell command in the background and end its turn. When the command
finishes, Claude Code starts a cycle of its own to continue.
> - Today the run ends when the turn ends. The command stops, and the
run fails with `native_finalization_missing`.
> - The Runner part of this work is complete in #21316 and the pull
requests below it: when a controller sends `providerStartedTurns` on
`turn.start`, the Runner keeps Claude while a background shell is
running, and gives the cycle of Claude to the controller as a turn that
the provider started.
> - Core reads Runner events through the Codex app-server driver, also
for ACPX agents. The driver refuses a turn that the controller did not
start, except a goal turn of Codex.
> - This pull request lets the driver accept a turn that the provider
started, only directly after a turn that the Runner marked as waiting.
> - The benefit is that the next parts can use that turn. Nothing
reaches this code until the last part sends the opt-in.

## Linked Issues or Issue Description

No public issue exists. The issue is described here.

This pull request is one of four parts on the Core side. They are
stacked on #21316:

1. The driver admits a turn that the provider started.
2. The consumer waits for that turn, for a limited time.
3. The server ends the wait when a person comments, and keeps the clock
across a restart.
4. The transport sends the opt-in. Only this part changes what a
production run does.

This is part 1. Its base branch is
`feat/adopt-claude-provider-started-turn`. Refs #19912 (design).

**What existing behavior does this improve?**

Nothing in production yet. What the driver does with a turn that the
provider started.

**Subsystem affected**

Paperclip Runner package, the controller side: `src/drivers/codex` (the
driver that Core uses to read a native session).

**Current behavior**

- A `turn/started` that the controller did not ask for fails the session
with `turn_binding_mismatch`, except for an active Codex goal.
- An interrupt with no active turn fails with "already terminal".

**Proposed behavior**

- A turn terminal that the Runner marked with `awaitingProviderTurn` and
that has no result puts the driver in a waiting state. The driver event
`turn.completed` carries the marker.
- In that state, a `turn/started` with `origin: "provider"` is admitted
the same way as a goal turn, and the driver event carries `origin:
"provider"`. In each other state it still fails closed.
- In that state with no active turn, an interrupt gives an
acknowledgement item with the status `provider_turn_wait_ended`. It
sends nothing to the Runner. The driver also gives that acknowledgement
when it gets the internal `paperclip/runTerminal` notification with a
private marker; a notification without the marker does nothing.
- An interrupt for a provider-started turn that is running is sent to
the Runner only when it names that turn.
- The driver saves the waiting state in the persisted session
(`awaitingProviderTurn`, written only during a wait). After a controller
restart, the restored session still admits the provider-started turn and
still acknowledges an interrupt during the wait. The driver restores the
state only when the session has no active turn, no stored result and at
least one finished turn, so a new run cannot take the wait of the run
before it.
- When a restart finds that the saved active turn is finished, the turn
read of the transport and `reconcile()` of the driver carry the same
marker as the live `turn/completed` notification.
`runnerd-codex-transport.ts` copies `awaitingProviderTurn: true` from
the saved `turn.completed` event to the turn object of a read.
- The driver ends the waiting state when the Runner reports the run
terminal.

**Reason and benefit**

The consumer in the next part needs these three signals: a marked
terminal, an admitted turn, and a clear end of the wait.

## What Changed

- `codex-session-terminal.ts`, `codex-session-notifications.ts`,
`codex-session-state.ts`, `codex-harness-session.ts`,
`codex-driver-values.ts`: 37 lines added, 4 removed.
- `codex-app-server-driver.provider-turn.test.ts` (new, 232 lines).
- `contracts/harness-driver.ts`, `contracts/native-session-backend.ts`,
`backends/harness-driver-backend.ts`, `codex-app-server-driver-impl.ts`:
the optional field in the two persisted shapes and its copy in each
direction.
- `live/runnerd-codex-transport.ts`: the marker on a recovered turn
terminal. The loop that builds the earlier terminals of a `thread/read`
is now the function `runnerdPriorTurnTerminals`, with the same logic, so
that a test can call it.

## State saved by the previous release

- One new optional field in the persisted driver session:
`awaitingProviderTurn`. A snapshot outside a wait has the shape of the
previous release.
- **State from the previous release, read by this release:** the field
is absent, so the driver does not wait. This is the behavior of today.
- **State from this release, read by the previous release:** the
previous reader copies the fields that it knows and ignores this one. No
reader rejects it, and the integrity digest covers the stored object as
written.

## Verification

At the head of this branch:

- `npx tsc --noEmit -p tsconfig.json`, `npx tsc --noEmit -p
tsconfig.test.json` and `pnpm run build:typescript` in
`packages/paperclip-runner`; `node
scripts/verify-runner-vendor-dependencies.mjs` and `pnpm typecheck` in
`server`; `node --test
scripts/__tests__/runner-release-crate-cache.test.mjs`: all passed.
- 6 driver tests: a provider-started turn after a marked terminal is
admitted and can report a result; a provider-started turn with no marked
terminal fails closed; an interrupt during the wait is acknowledged; a
running provider-started turn is not interrupted by the end of the wait;
the marked notification ends the wait and a notification with no marker
does not; a normal terminal is not changed.
- Restart tests: a session restored from a snapshot with the waiting
state admits the provider-started turn and settles its result; a
snapshot outside a wait has no new field; a restored session with no
waiting state refuses a provider-started turn as today; an interrupt
during a restored wait is acknowledged. A saved active turn that the
read reports finished with the marker gives a `turn.completed` with
`awaitingProviderTurn: true`; with no marker it is an ordinary terminal.
- `harness-driver-backend.test.ts`: the field goes through a serialized
checkpoint only during a wait. `runnerd-codex-transport.test.ts` (324
passed): the marker is on a recovered terminal only for a
`turn.completed` with a boolean `true`.
- `src/native-session-runtime.test.ts`: 297 passed, with the real
`paperclip-runnerd`.
- Mutation checks for the restart work, each made a named test fail: no
restore of the state; no copy in the backend; no copy in the transport;
no marker argument in `reconcile()`; no end of the wait at the run
terminal.
- Mutation checks: the acknowledgement branch removed, the admission
removed, and the marker check removed each made a test fail.

Not run: the full suite. CI supplies that coverage.

## Risks

- **No production effect.** The Runner sends the marker only after the
opt-in of part 4.
- **The admission rule is shared with Codex goal turns.** The existing
goal tests pass with no change.
- **Known limit after a controller restart.** If the Runner already runs
a provider-started turn and the saved active turn is the turn before it,
recovery fails closed, as today for each unexpected active turn.
- **A controller turn that starts during a wait** clears the waiting
state when its `turn/started` arrives, not when it is sent. A snapshot
in that short time can hold an active turn and the field together; the
restore rule drops the field.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`) in Claude Code planned the
change and the split, reviewed the diff, and ran a part of the
verification again.
- Claude Opus 5.5 (`claude-opus-5-5`), as a Claude Code subagent with
tool use (shell, file edits, local test runs), wrote the change and the
tests and cut this slice from the full change.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-13 20:45:49 -08:00
Dotta ca7d42ba71 fix(heartbeat): refresh queued dispatch when task drains end (#21562)
## Thinking Path

> - Paperclip runs agent work for a company.
> - A task drain holds new run starts while maintenance or idle handoff
is in progress.
> - A queued dispatcher can keep an unreconciled request and a 30-second
retry while the drain is active.
> - Ending the drain can resume domain work without refreshing that
dispatcher's saved deadline.
> - This change refreshes the pending queued owner when a drain ends or
its expiry is observed.
> - Normal admission checks and real retry deadlines still apply.

## Linked Issues or Issue Description

Related: #20640. That change refreshes deferred receipt workers. This
change refreshes the separate queued-run dispatcher.

**What happened?**

A queued-work notification received during a drain could leave a
30-second fallback in the dispatcher after the drain ended. The owner
stayed unreconciled after useful work had settled. This delayed its next
pass and the instance's idle state.

**Expected behavior**

Refresh an unreconciled queued owner when normal dispatch becomes
allowed. Keep other drain holds, admission rules and scheduled retry
deadlines.

**Steps to reproduce**

1. Let the normal dispatcher complete its first pass.
2. Start an idle task drain and commit a company-scoped queued-work
notification.
3. Stop the drain, or observe its expiry, before the saved fallback is
due.
4. Check that the scoped queued pass runs before that fallback. The two
regression cases fail on the unchanged owner and pass with the fix.

**Paperclip version or commit**

The main-specific patch is based on
`f7b1651799d8c0c660a76266878c420a12f0beca`. The bug was found in
assembled sleep acceptance. The affected queued-dispatch, drain and
notification modules match main.

## What Changed

- Listen for normal drain changes and observed expiry in the queued-run
owner.
- Wake that owner only when dispatch is allowed and its reconciliation
is pending.
- Forward the deadline worker's optional stop callback after its stop
fence.
- Remove both subscriptions when the owner or coordinator stops.
- Add release, expiry, remaining-gate, shutdown, retry-deadline and idle
controls.

## Verification

Local checks used the integrated feature branch that exposed the issue.
The main-specific port preserves main's existing `canRun` logic and
excludes the private branch's extra cancellation hook.

- Both new release and expiry cases failed at the intended scoped-resume
assertion before the product change.
- Six dispatch cases pass with the normal package import.
- Fourteen existing deadline controls and nineteen queue-order controls
pass. Their inputs did not change during the test-import correction.
- The complete normal Server typecheck and its three affected-file
program checks pass.
- The first typecheck caught a test import outside the Server root. The
import now uses the existing public package wildcard. The corrected
six-case suite and normal typecheck pass.
- Normal Source SDK input verification passes.
- Exact-head CI at `fd623d3ba9345e5ec8f252c55932ff797cd57e8f` completed:
54 passing checks, four expected skips and one neutral result; Snyk
passed. Greptile reviewed this exact head at 5/5 with no actionable
findings. All review threads were read; none remain open. A full local
repository build and suite were not run in this checkout.

The existing admission-drain contract in `doc/execution-semantics.md`
already requires prompt admission after drain release. This fix restores
that contract. No configuration or command documentation changes are
needed.

## Risks

Drain events can request an earlier pass for a pending owner. The normal
permission, reconciliation, coalescing and earliest-time checks remain
active. The scheduled retry owner has no new listener. Shutdown must
remove both listeners; the coordinator-stop control covers that path.

This PR does not claim deployed suspension or complete transport sleep
acceptance.

## Model Used

OpenAI `gpt-6.1-sol` assisted with implementation and independent
review. The Codex managing agent performed technical review and
verification. The host does not expose its exact runtime model ID or
context-window size. Tools used include source inspection, code edits
and local test execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes (not
required: the existing contract remains accurate)
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge
2026-11-13 22:45:26 -06:00
DottaandClaude Opus 5.5 9824277f65 perf(run-events): append events whose only wake is chat progress in one statement (#21559)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Most runner events that are not text deltas
(`tool.execution.started/progressed/completed`, `item.completed`) wake
the chat progress projection
> - That wake is registered inside the append's transaction, so each
such event takes 4 statements (BEGIN, the run lock read, the append,
COMMIT), where an event without a wake takes 1 (#21511)
> - The chat progress projection rebuilds itself from the stored events;
the wake only tells it when to look
> - This pull request writes an event whose only wake is chat progress
with the one-statement append, and sends the wake after that statement
commits
> - The benefit is 1 statement instead of 4 for most non-delta runner
events, with response, reassessment, retention and recovery wakes
unchanged

## Linked Issues or Issue Description

Refs #21511 (one-statement append). Refs #21528 (lock read returns the
transaction id). Refs #21549 (load model at recorded rates).

**What existing behavior does this improve?**
`appendHeartbeatRunEvent` for native source events whose only wake is
chat progress.

**Subsystem affected**
Server run event ingest (`heartbeat-run-events.ts`), the chat progress
notification helper, and two additive exports in
`packages/db/src/work-signals.ts`.

**Current behavior**
A progress event opens a transaction, reads the run lock with
`runtime_mode`, the bound task and the transaction id, probes for a
replay, registers the chat-progress wake as a tracked intent, appends,
and commits.

**Proposed behavior**
When chat progress is the event's only wake, the append takes the
one-statement path of #21511. That statement also returns the run's
`runtime_mode` and bound task id from the row it locks, and, where
commit notifications are on, calls `pg_notify` inside the same statement
for runs that are native and bound to a task, so other server processes
get the wake at commit. After the statement returns, this process
publishes the wake as settled to its own listeners. A replay, a gap, a
changed run or any other fallback takes the existing transaction, which
registers the wake as before.

**Reason and benefit**
At recorded rates these events are about 61% of the non-delta events
(#21549), so this cuts about 3 statements per such event.

**Breaking changes**
None. No schema change.

## What Changed

- `appendHeartbeatRunEvent`: `progressOnly` (progress set, every other
wake clear) also qualifies for the one-statement path.
- `APPEND_NEW_NATIVE_SOURCE_EVENTS_WITH_PROGRESS_SQL`: the one-statement
append, plus `runtime_mode` and the bound task id (`boundRunIssueIdSql`,
rendered once) from the locked row, plus a conditional `pg_notify`.
- `chat-native-progress-notifications.ts`:
`chatNativeProgressCommitNotification` and
`publishChatNativeProgressSettled`.
- `work-signals.ts` (additive): `databaseWorkCommitNotification(owner,
topic)` returns the cross-process payload when notifications are on;
`publishDatabaseWorkSettled(owner, topic, hint)` publishes a settled
wake to this process's listeners.

## Verification

- Red/green on embedded PostgreSQL. Each new test failed against
`origin/master` first:
- `heartbeat-run-event-atomic-append.test.ts` "appends an event whose
only wake is chat progress in one statement and wakes progress after it
commits": 1 statement (before: a transaction), the wake is published
once as settled with no intent, and a replay publishes nothing.
- "keeps the transaction for the last events of a turn and a run, and
one statement for progress-only events":
`tool.execution.started/progressed/completed` and a non-final
`item.completed` take one statement; a final `item.completed`,
`run.result.accepted`, `turn.completed`, `turn.failed`, `turn.cancelled`
and `turn.interrupted` keep the transaction.
- "sends the progress wake to other server processes with the one
statement's commit": with commit notifications on, another connection
that listens on the work channel receives exactly one chat-progress
notification, and a replay sends none.
- `chat-native-progress-work.test.ts` "wakes the projection from a
one-statement progress event after the statement commits": the real chat
progress worker publishes the progress for an event appended in one
statement.
- Mutation check: without the settled publish after the statement, that
test fails.
- Lost-wake repair (condition b): an event row stored without any wake
models a lost hint.
- "repairs a lost progress wake with the run's next progress event":
nothing is published for the stored event; the run's next progress event
publishes the progress.
- "repairs a lost progress wake with the startup reconcile": a worker
that starts after the stored event publishes it.
- The existing response-wake test now uses a final reply and still shows
the intent before the write, and the race test of #21528 uses a final
reply so it still covers the transaction path.
- Other focused suites, one file at a time: the port suite with
conformance (105), `native-observer-replay` (1),
`paperclip-semantic-conformance` (1), `native-interaction-bridge` (7),
`heartbeat-run-event-sequencing` (5), `native-event-boundary-golden`
(1), `heartbeat-native-restart-suspension` (115), `live-recovery-work`
(51), `work-signals` (25), atomic-append (31),
`chat-native-progress-work` (38). `pnpm --filter @paperclipai/server run
typecheck` and `pnpm check:module-boundaries` pass.

- CPU per operation (a local script, not committed: 800
`tool.execution.progressed` appends to one native task run on embedded
PostgreSQL, one connection, three rounds, `origin/master` code against
this branch): main-thread CPU per event 0.28–0.50 ms before and 0.10 ms
after in the warm rounds; wall time per event 0.84–1.04 ms before and
0.52–0.53 ms after.
- Inside a caller's transaction: new test "keeps a progress-only event
inside a caller's transaction on the locked path, and settles its wake
only when that transaction ends". `publishDatabaseWorkSettled` publishes
at once only outside a transaction; inside one it waits for the
outermost commit and drops the wake on rollback (new
`work-signals.test.ts` test, which failed with the unguarded publish).
The one-statement path is only taken when the owner is not a transaction
(`!is(db, PgTransaction)`, #21511), so the two guards agree.

## Risks

- **Lost wake.** If the statement's outcome is unknown (the connection
drops after the server ran it), this process sends no wake for that
event. In a chat channel (Slack, Telegram) the progress line for that
task can then stay one step behind (for example it does not move from
"working" to the latest tool step) until the run's next runner event, a
run milestone, or a server restart. A turn normally ends with an event
that keeps the transaction
(`turn.completed/failed/cancelled/interrupted`), and the run's
completion notifies chat run milestones, which include chat progress
(`notifyChatRunMilestones` in `run-completion.ts` and
`native-run-finalizer.ts`). So progress cannot stay stale after a run
ends; in the worst case it lags one step for the rest of a turn. The
projection rebuilds from stored events, so nothing is lost or
duplicated.
- The chat-progress wake no longer publishes an intent before the write.
Intents let the coordinator fence a pass against an in-flight write; the
settled wake after the commit makes the projection run again after the
row is visible.
- `packages/db/src/work-signals.ts` gets two additive exports and no
change to existing behavior.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:37:35 -06:00
DottaandClaude Opus 5.5 8c0c7c6abb perf(ui): load the @task mention pool after the task page settles; add a cold-load perf spec (#21557)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - People often open a task by its URL, for example from a link in a
message or after a reload
> - #20592 made a hard load of a task URL read the chat's first frame
from one `chat-bootstrap` request
> - The task page still sends a large read as it mounts: 100 full
recently updated tasks for the `@task` mention picker. On staging this
read returned 395 KB and took 3.9 s of server time
> - The hosted database is CPU-bound, so reads that do not affect the
first frame must not compete with the reads that do
> - Also, no test measured a cold direct load under realistic latency or
guarded the one-bootstrap request shape
> - This pull request starts the mention pool after the task page
settles. It also adds a cold-load perf spec with a delay on each API
request and an always-on request-shape test
> - The benefit is less concurrent server work during a cold task open,
and a regression guard for the one-round-trip first frame

## Linked Issues or Issue Description

Refs #20592 (hard-load chat bootstrap that this pull request measures
and guards). Refs #20355 (the `chat-bootstrap` endpoints).

**What existing behavior does this improve?**
The task page loads the `@task` mention pool when it mounts.

**Subsystem affected**
Board UI task page (`ui/src/pages/IssueDetail.tsx`) and the issue-detail
perf harness (`tests/perf/issue-detail`).

**Current behavior**
`GET /api/companies/:id/issues?limit=100&sortField=updated&sortDir=desc`
starts with the chat's first-frame reads. It starts before the first
chat frame in 14 of 14 local cold loads. The picker needs it only after
the person types `@`.

**Proposed behavior**
The read starts after the page has painted and its own reads have
finished. On a page that keeps reading, it starts after at most 5 s.

**Reason and benefit**
The read never affects the first frame. Deferring it removes a large
concurrent read from the busiest part of a cold open.

**Breaking changes**
None. If the person types `@task` in the first moments after a cold
load, the task suggestions can appear about half a second later.

## What Changed

- `ui/src/hooks/usePageSettled.ts`: new hook. It returns `true` once
`whenPageSettles` runs (the page has painted, the tab is visible, no
query is fetching, and the browser is idle), or after
`PAGE_SETTLED_MAX_WAIT_MS` (5 s). The cap keeps a long-running read from
keeping the picker empty.
- `ui/src/pages/IssueDetail.tsx`: the mention-pool query is enabled only
when `usePageSettled()` is true.
- `ui/src/hooks/usePageSettled.test.tsx`: tests that the hook waits for
in-flight reads and that the cap applies.
- `ui/src/pages/IssueDetail.test.tsx`: mocks the hook to `true`, so
existing tests keep their behavior.
- `tests/perf/issue-detail/cold-bootstrap.perf.spec.ts`: new spec. It
delays every API response (300 ms by default) and opens the task URL in
a new browser context, with the JavaScript in the HTTP cache and no app
storage. The finish event is the first readable frame: the history is
revealed, the latest saved message is visible and the composer is
editable.
- The request-shape test always runs. It requires one `chat-bootstrap`
request and no per-endpoint first-frame read before the frame. Reads
that start after the live socket is created are excluded, because they
are its catch-up refresh.
- The benchmark runs only when `PAPERCLIP_ISSUE_PERF_COLD_RUNS` is set.
It writes the median, the spread, the request count before the frame,
the last requests before the frame, the JavaScript bytes transferred and
a request timeline.
- The delay uses CDP `Fetch` interception instead of Playwright routing,
because routing disables the HTTP cache.
- `tests/perf/issue-detail/README.md`: documents the spec.
- My earlier draft also added a deadline for a stalled bootstrap. Master
already has one (`CHAT_BOOTSTRAP_SHARE_DEADLINE_MS`, 5 s), so this pull
request does not include it.

## Verification

Measurement setup:
- Production UI builds of master `4a78124a3e` and of this branch, served
by one local server (`PAPERCLIP_UI_DEV_MIDDLEWARE=false`, embedded
PostgreSQL).
- A delay of 300 ms on each API request.
- Cold direct task URL with 24 saved comments and a plan.
- Master and the branch alternated: 2 rounds of 7 runs each.
- The host was shared with other agent sessions.

| Build | Runs | First readable frame, median | p25–p75 | Min–max | API
requests before frame (median) | Mention pool started before frame |
|---|---:|---:|---:|---:|---:|---:|
| master | 14 | 1,422 ms | 1,336–1,525 | 1,284–7,398 | 50 | 14 / 14 |
| this branch | 14 | 1,390 ms | 1,338–1,457 | 1,272–2,207 | 49 | 0 / 14
|

- The spec adds the API delay through CDP request interception, so the
HTTP cache stays on. The measured loads downloaded 0 KiB of JavaScript
in every run. The spec records this as `jsTransferBytes`. Each build has
one outlier (7,398 ms on master, 2,207 ms on this branch), from host
contention.
- With this branch, the mention pool starts after the frame in every run
(350–400 ms after it in an earlier set).
- The first-frame timing is neutral locally. The local database answers
the pool read in milliseconds, so it does not compete with other reads.
The intended benefit is lower concurrent server load on a busy database.
This pull request does not show it as a local latency gain.
- On master and on this branch, the hard load adopts the bootstrap and
the frame is no longer held by task reads. The frame now comes right
after the shell reads return: `auth/get-session` and
`instance/settings/experimental`, then `companies` and
`workspace-preferences/me`. Profiling marks confirm this: the chat
thread mounts after `workspace-preferences/me` returns, and all
readiness gates are ready within 30 ms.
- The request-shape test passed 5 of 5 on master and 5 of 5 on this
branch.
- `layout-stability.spec.ts`, `plan-startup.spec.ts` and
`cold-bootstrap.perf.spec.ts` against this branch's build: 18 passed.
- `ui` Vitest for `usePageSettled.test.tsx` and `IssueDetail.test.tsx`:
278 passed. The `ui` typecheck passed. `pnpm check:token-gates` passed.
- `pnpm verify` exits 2 because `tests/perf/issue-detail/README.md` is a
global CI input. Required CI supplies the broad coverage.

To reproduce the measurement:

```sh
PAPERCLIP_ISSUE_PERF_BASE_URL=http://127.0.0.1:3287 PAPERCLIP_ISSUE_PERF_COLD_RUNS=7 PAPERCLIP_ISSUE_PERF_LABEL=branch \
  pnpm exec playwright test --config tests/perf/issue-detail/playwright.config.ts cold-bootstrap -g benchmark
```

## Risks

- Low risk. The only product change is when one read starts.
- If the page keeps a query in flight, the pool waits up to 5 s. During
that time, `@task` suggestions list only agents, projects and members.
- The request-shape test treats reads that start after the live socket
is created as catch-up refreshes. If the socket connects before the task
page mounts, such a test would not see a first-frame read that started
later. The test still requires exactly one bootstrap request before the
frame.
- The perf spec is not in the default CI browser suites. It runs when
the issue-detail harness runs.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used in Claude Code
with tool use (shell, file editing, Playwright browser runs). It wrote
the change, the tests and the measurements.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
canary/v2026.1114.0-canary.2
2026-11-13 22:30:03 -06:00
DottaandClaude Opus 5.5 54c3616ad9 fix(recovery): ignore the non-assignee comment wake marker in the failed-preparation proof (#21550)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When a Runner run fails in setup before any provider work, a
person's Retry must start a new run
> - Admission proves "no provider work" for a failed preparation by
checking that the run record holds only known, harmless fields
> - The run profile allowlist did not include
`aiConnectionNonAssigneeCommentWake`, a boolean that dispatch writes
when it admits a run from a non-assignee comment
> - So a run with that marker never passed the proof. Its Retry waited
on `previous_execution_unverified` forever, even after saved-work
recovery finished and the workspace was handed on unchanged
> - This pull request accepts the marker (boolean only) in the proof,
and in the related workspace-wait parent proof
> - The benefit is that these tasks get a working Retry, and a Retry
that was saved earlier starts by itself once the proof passes

## Linked Issues or Issue Description

Refs #21481, #21497.

**What happened?**
On a hosted staging instance, a Runner run failed in setup ("RPC call
environmentExecute timed out") while it prepared. Saved-work recovery
finished ("Saved work recovered."). The workspace was handed on
unchanged for this run. A person pressed Retry two times. Both Retries
waited on `previous_execution_unverified`, and no run started.

**Expected behavior**
A run that failed in setup before any provider work
(`executionRecovery.providerWorkStarted: false`, no process, no
execution events, stopped sandbox) gets a working Retry. A saved Retry
starts by itself when its gate clears.

**Steps to reproduce**
1. Let dispatch admit a Runner run from a non-assignee comment, so the
run profile has `aiConnectionNonAssigneeCommentWake: true`.
2. Make the run fail in setup during preparation.
3. Press Retry. See that the request waits on
`previous_execution_unverified`.

**Paperclip version or commit**
Observed on staging at `e82c4f0a31`; still present on `master` at
`6b631dd477`.

## What Changed

- `server/src/services/cancelled-native-startup.ts`: the
failed-preparation proof accepts `aiConnectionNonAssigneeCommentWake`
when its value is a boolean. The workspace-wait parent proof ignores the
same key.
- `server/src/services/legacy-preparation-retry-hold.test.ts`:
staging-shape tests for a recovered setup failure (failed restore, a
workspace that still names its location, the marker, all together), and
a Retry that attaches to an earlier refused request and then starts.

## Verification

- `cd server && npx vitest run
src/services/legacy-preparation-retry-hold.test.ts`: 12 passed. Without
the fix, the `comment_wake_marker`, `all` and `attached_after_refusal`
cases fail (red), and the other cases pass.
- Focused test file only (the machine was loaded). CI supplies
typecheck, build and the full suite.

## Risks

- Low risk. The marker only records why dispatch admitted the run. Every
other proof clause (no process, no execution events, no native result,
stopped leases, settled preparation evidence, workspace repair) is
unchanged. A non-boolean value still fails the proof.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), via Claude Code with tool use
(shell, file edits, test execution).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:02:55 -06:00
DottaandClaude Opus 5.5 6b9e45787f fix(recovery): continue a task whose system retry was refused at claim behind a saved-work hold (#21505)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When a run stops before it finishes, automatic recovery continues
the task, through the fresh-continuation fallback, in a new run that
starts from the saved work
> - A follow-up run (for example one started by a new comment or by
finished subtasks) can fail before any provider work and leave a
saved-work hold. The system then queues its own retry, and the dispatch
gate cancels that retry at claim with
`execution_reconciliation_required` because the hold is still in place
> - The fallback takes over the hold when that happens, but it continued
only "recent interruptions". That test required the failed run to be a
recovery run itself or to fail with a restart code. A first follow-up
run that failed with `adapter_failed` passed neither test, even though
the system had already decided to retry it. So the fallback answered "a
person must press Retry". The backlog pass used the same test, and the
assigned-work sweep never listed the task. The task stayed blocked with
no run, no pending path and no Retry button
> - This pull request treats a system retry refused at dispatch as proof
that the system continues the task. It also stops cancelled-at-dispatch
attempts from counting against the hourly budget, lets the sweep
continue a blocked task whose latest run was refused at dispatch, and
keeps recovery notices from asking for a Retry that the task does not
offer
> - The benefit is that such a task continues by itself in exactly one
fresh run from the saved work, whether the saved work was recovered
before or after the refusal, and a person's own failed Retry keeps
today's behaviour

## Linked Issues or Issue Description

Refs #21394, #21348, #21363, #21478

**What happened?**
A task's follow-up run, a legacy run that was not a recovery run, failed
before any provider work with `adapter_failed`. Its files could not be
restored, so it left a saved-work hold
(`legacy_execution_requires_reconciliation`, `workspaceRestoreFailure:
restore_failed`). The system queued a retry of that run. The dispatch
gate cancelled the retry at claim with
`execution_reconciliation_required`, and the "older run hold" path
handed the hold to the fresh-continuation fallback
(`run-dispatch/adapters/postgres.ts` `settleOlderRunHold`). The fallback
found the refused run and asked `interruptionEligibility` whether the
interruption was recent. That predicate required
`continuesAutomaticallySql(source)` (a recovery run, or a restart,
deploy or bootstrap code), so it returned `person_owned` /
`retry_needed`. The stranded reconciler then blocked the task with
"Press Retry", but the saved-work hold reports that Retry is not
available. This happened twice in a local restart-chaos wave. In one
case the saved work was recovered two minutes before the claim. In the
other it was recovered thirty seconds after the claim.

**Expected behavior**
A task never rests blocked with no live run, no pending path and no
Retry. While the saved-work recovery runs, that recovery is the pending
path. When it finishes, one continuation starts from the saved work
through the fallback (isolation, the shared hourly cap). A run that was
cancelled at dispatch before it started does not use up the automatic
budget. A person's own failed Retry is still left for that person.

**Steps to reproduce**
1. Fail a follow-up run (wake reason `issue_children_completed`) before
provider work with `adapter_failed` and `workspaceRestoreFailure:
restore_failed`.
2. Queue the system's retry of that run (`retryOfRunId`, wake reason
`issue_continuation_needed`) and claim it. The dispatch gate cancels it
with `execution_reconciliation_required`.
3. Before this change, no continuation is recorded, the stranded
reconciler blocks the task, and nothing ever continues it.

**Paperclip version or commit**
master at 2756be2f32

## What Changed

- `fresh-continuation-fallback.ts`, `interruptionEligibility`: the
interruption counts as one the system continues when the failed run's
cause is system-owned and a later run that the dispatch gate refused
before it started is a system recovery run (`recoveryRunSql`, which
excludes a person's Retry). A failed run that was itself a person's
Retry (`task_retry` or `explicitUserContinuation`) never qualifies this
way. The 24-hour recency bound is unchanged. The claim-gate hand-off,
the backlog pass and its deadline query all use this one predicate.
- `recovery/continuation-ledger.ts`,
`recovery/task-automatic-attempts.ts`: one shared rule,
`startedOrPendingAutomaticRetry`. An automatic retry that the dispatch
gate cancelled before it started (`cancelled`, no `startedAt`,
`execution_reconciliation_required`) does not count toward the hourly
cap in either ledger.
- `recovery/assigned-work-sweep.ts`:
- When the latest run was refused at dispatch, the sweep continues from
the run it was refused for through the fallback. Before, it gave up.
- A blocked task with an active hold and a recent refused-at-dispatch
latest run is now listed as a suspect. The check hands it to the
fallback before it answers `left_to_hold_owner`. A pending recovery
becomes `waiting_on_recovery` with its recheck time.
- Holds that wait for a person's configuration or confirmation are still
excluded.
- Review follow-up (`recovery/task-automatic-attempts.ts`,
`recovery/assigned-work-sweep.ts`):
- The task ledger counts a sweep's fresh-continuation record (`trigger:
assigned_work`) itself. It no longer depends on a companion activity
entry, and an activity entry marked `continuedFromSavedWork` is not
counted a second time. Every sweep continuation counts once, whatever
path recorded it.
- When the sweep claims a continuation, it marks its claim entry
`continuedFromSavedWork` inside the transaction that records the fresh
continuation. A restart between the two writes can no longer leave one
attempt counted twice.
- The sweep's continuation behind a refused run spends the per-minute
repair budget (`ASSIGNED_WORK_REPAIRS_PER_MINUTE`). The excess is
deferred with a `retryAt`.
- `recovery/stranded-notice.ts`, `recovery/service.ts`: when the task's
execution blocker offers no Retry, the escalation notice and the
recovery action's next action leave out every Retry instruction. They
say that Paperclip is holding the task until its saved work is settled.
- No change to the dispatch gate (`heartbeat/queue.ts`, the restore-hold
predicate in `run-dispatch/adapters/postgres.ts`, or
`execution-blocker.ts`). The recorded continuation settles the hold, so
the delivered run is not refused again.

## Verification

Red before the change, green after (embedded Postgres, real services:
`terminalizeLegacyExecution`, the dispatch adapter's
`cancelStaleQueuedRun`, `reconcileStrandedAssignedIssues`,
`deliverReconciledExecutions`, `continueHeldExecutions`,
`recoveryService().assignedWork`):
- `server/src/services/fresh-continuation-fallback.test.ts`
- "continues a follow-up … when the saved work was recovered before the
system retry was refused at claim": red (`continuationDelivery:
"invalidated"`, no continuation). Green: the continuation is recorded at
the claim, and the stranded reconciler does not block the task. The
delivered run is queued and passes the claim gate (`not_stale`).
- "… recovered after the system retry was refused at claim": red (no
pending check). Green: the claim records `exact_recovery_pending` /
`saved_work_recovery_running`. When the recovery completes, the backlog
pass continues the task, and the delivered run passes the claim gate.
- "continues a follow-up … with its files in place, after its system
retry was refused at claim": red, then green. This is the control
without a restore failure.
- "still leaves a person's failed Retry of a follow-up for that person
…": stays `person_owned` / `retry_needed`, before and after.
- "counts only automatic retries that started, or still may, against the
task's hourly budget": new.
- `server/src/__tests__/failed-continuation-continues-again.test.ts`,
"the sweep continues a blocked task whose system retry was refused at
claim behind a recovered saved-work hold": red (the task was not a
suspect). Green: `repaired`, a pending continuation in the hold, the
task back in `todo`, and no wake enqueued. The next check reports a
pending intent.
- Same file, "a continuation the sweep records behind a refused run
counts once toward the task's hourly cap": red (5 counted after 5
attempts plus the new continuation), green (6, and the transient-retry
cap check refuses the next attempt).
- Same file, "continuations behind refused runs spend the sweep's
per-minute repair budget": 20 tasks. Red (20 continued in one pass),
green (10 continue, 10 are deferred with a `retryAt`, and the 10
continue in the next pass).
- Same file, "a sweep continuation interrupted right after it was
recorded still counts once": the continuation commits, then the
post-commit activity write fails (a simulated server stop). Red: 6
counted for 5 real attempts. Green: 5 counted, and the next automatic
retry is still allowed.
- `server/src/services/recovery/stranded-notice.test.ts`, "never tells a
person to press Retry when the task offers no Retry": new.

Local runs on this branch, one file at a time:
- `fresh-continuation-fallback.test.ts`: 77 passed. The suite's
`afterAll` teardown hit the 30 s hook timeout on a heavily loaded
machine.
- `failed-continuation-continues-again.test.ts`: 32 passed
- `assigned-work-invariant.test.ts`: 31 passed
- `stranded-notice.test.ts`: 33 passed
- CI on the first head (ce4431410b): every required check passed
- `server` `tsc --noEmit`: clean
- `pnpm verify --explain` selects most of the repository (2666 files),
so required CI supplies the broad coverage.

## Risks

- More tasks continue automatically. Each continuation goes through the
existing fallback. The 24-hour recency bound, the per-task hourly cap
(6), isolation of the next run, and every person-owned gate
(person-owned failure codes, a person's Stop, review, a person-owned
hold, a person's failed Retry) are unchanged.
- The hourly cap now ignores automatic retries that never started. A
task whose retries are refused at dispatch again and again is still
bounded: the refusal hands the hold to the fallback, and the fallback
records one continuation per source run.
- Known separate gap, not changed here: after an interrupted
preparation's holds are resolved (`settleInterruptedNativeBootstrap`),
`scheduleBoundedRetry` can report `exhausted` or `blocked`, and nothing
follows that up. The assigned-work sweep may now pick such a task up
through the refused-at-dispatch path, but this pull request does not
prove that.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, local test runs)

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 22:01:39 -06:00
Nicky LeachandClaude Fable 5.2 f7b1651799 feat(acpx-sidecar): adopt Claude's own cycle as a provider-started turn when the Runner arms it (#21316)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Paperclip Runner runs Claude Code through the ACPX sidecar.
> - Claude can start a shell command in the background and end its turn.
When the command finishes, Claude Code starts a cycle of its own to
continue. Today the Runner stops Claude at the end of the turn, so the
command stops and the run fails with `native_finalization_missing`.
> - A stack of pull requests lets the Runner accept that cycle as a new
turn of the same run, when a controller asks for it.
> - The parts before this one give the adapter an ordered stream of
those cycles (#21421), declare the contract (#21422), and make
`paperclip-runnerd` wait for such a turn (#21423).
> - This pull request is the last part in the Runner. The sidecar reads
the stream, and gives a cycle to `paperclip-runnerd` as a
provider-started turn only when `paperclip-runnerd` armed it.
> - The benefit is that a later change in Core can let a run continue
when background work finishes. Until Core sends the opt-in, a production
run does the same as today.

## Linked Issues or Issue Description

No public issue exists. The issue is described here.

This pull request was first opened with the full change (49 files). It
is now split. The parts, in stack order after #21311:

1. #21419: sidecar preparation (a refactor).
2. #21420: converter for raw ACP session updates.
3. #21421: cycle stream at the adapter.
4. #21422: contract for provider-started turns.
5. #21423: `paperclip-runnerd` waits for a provider-started turn.
6. This pull request: the sidecar adopts a cycle.

Its base branch is `feat/runnerd-await-provider-turn` (#21423). The code
at the head of this branch is the same as before the split; only the
base changed. Review comments made before the split are about code that
is now in the parts above. Refs #19912 (design).

**What existing behavior does this improve?**

What the ACPX sidecar does with a cycle that Claude starts by itself.

**Subsystem affected**

Paperclip Runner: `src/cli/acpx-runtime-sidecar.ts`. Claude only.

**Current behavior**

- The sidecar does not read the cycle stream of the adapter (#21421), so
each cycle is declined and the background state becomes unknown.
- The sidecar does not know the command `session.provider_turn`, so
`paperclip-runnerd` (#21423) never waits.

**Proposed behavior**

- The command `session.provider_turn` with `armed: true` arms the
sidecar for one turn. The sidecar accepts it only for the terminal that
reported a running background shell, with no active turn and a provider
that is alive. One flag holds the arming. It is cleared at each turn
start and turn end, on suspend and close, and when the provider is lost.
- One loop, `consumeProviderCycles`, reads the stream. For each cycle:
- if the sidecar is armed and no turn is active, the sidecar adopts the
cycle: it emits `runtime.turn_started` with `{ "origin": "provider" }`,
then the events, the usage and the terminal of the cycle, as for a
normal turn;
- if the cycle cannot be decided yet (the prompt turn is still settling,
or `paperclip-runnerd` did not arm yet), the sidecar holds it for a
maximum of 5 seconds, and then adopts or drops it;
- in each other case the sidecar drops the cycle. The stream ends, and
the background state becomes unknown, so `paperclip-runnerd` restarts
the provider at the end of the turn as today.
- A Paperclip tool call or a permission request of a held cycle waits
for the decision of its own cycle. It is never bound to the prompt turn
or to a different cycle.
- If an adopted cycle proposes a plan (`ExitPlanMode`), the sidecar
publishes the same `plan.proposed` event as for a controller turn, and
the turn ends with the same terminal: completed, with no
`backgroundTasks`, so `paperclip-runnerd` does not wait again. A
controller turn is not changed.
- If the provider exits while the sidecar is armed with no turn active,
the sidecar sends a `runtime.diagnostic` with the code
`ACPX_PROVIDER_PROCESS_LOST`, and `paperclip-runnerd` ends the wait.
- `turn.cancel` for a provider-started turn cancels the cycle and closes
the provider.
- With no arming, nothing is shown. No controller sends the opt-in yet,
so `paperclip-runnerd` never arms.

**Reason and benefit**

- Only `paperclip-runnerd` decides that a cycle becomes a turn. The
sidecar cannot start a turn by itself.
- One reader of one ordered stream means that two cycles cannot overlap.

## What Changed

- `src/cli/acpx-runtime-sidecar.ts` (127 lines added): the command, the
arming flag, `consumeProviderCycles`, `adoptProviderCycle`,
`pumpProviderTurn`, and the wait of `waitForTool` for a cycle that is
not decided.
- `src/cli/acpx-runtime-sidecar.provider-turn.test.ts` (new, 320 lines)
and two fake hosts in `src/test-support` (227 lines). One host follows a
script. The second runs the real cycle stream and the real background
task tracker.
- `src/cli/acpx-runtime-sidecar.test.ts` (2 lines): one more parameter
for a test harness.

No schema, no Rust and no persisted state changes in this part.

## Verification

At the head of this branch, from `packages/paperclip-runner`:

- `pnpm run typecheck:typescript`, `npx tsc --noEmit -p
tsconfig.test.json` and `pnpm run build:typescript`: no errors. The
server check `node scripts/verify-runner-vendor-dependencies.mjs`, the
cache guard and the contract tests: passed.
- `npx vitest run src/cli src/drivers/acpx
src/live/runnerd-acpx-provider-turn-wait.test.ts
src/drivers/runner-tool-bridge.test.ts
src/native-session-runtime.provider-turn.test.ts`: `128 passed (128)`
files, `2147 passed`, 9 skipped.
- `cargo test -p paperclip-runner-core --lib` and the `acpx_*` suites:
all passed. No Rust changed in this part.
- The sidecar tests of this part: a cycle that starts before the arming
command, or while the prompt turn settles, is held and then adopted with
its tool call; a cycle is never shown when `paperclip-runnerd` did not
arm, disarmed, armed for a different terminal, armed for a quiet
terminal, or armed after the provider exited; two cycles in sequence
with `{ live: 2 }`, then `{ live: 1 }`, then an unknown state; cancel; a
lost provider; Codex gets no change.
- Mutation checks, each made the named tests fail: the sidecar arms with
no matching terminal; no hold; the old hold condition that drops a cycle
while the prompt turn settles; no wait of a tool call for the decision.
- Live run with a real Claude account, not in CI, with the real ACP
runtime and the real Anthropic API (model `claude-sonnet-5-5`), at the
adapter level: one background shell gave `{ live: 1 }`, one cycle with
its own usage, then an unknown state. Two shells of 8 s and 28 s gave `{
live: 2 }`, `{ live: 1 }`, then unknown. All Paperclip tool calls were
admitted, and none was refused.

Not run: a live run through the real sidecar process,
`paperclip-runnerd` and Core (that path needs an API key, not a local
sign-in); Linux; `pnpm verify --full`. CI supplies the broad coverage.

## Risks

- **The feature is off.** No controller sends the opt-in, so the sidecar
is never armed. The new code that runs today is the loop that reads the
stream: with no arming it drops each cycle, and the background state
becomes unknown as before.
- **A follow-up for a command that finishes while Claude reports on an
earlier one is lost.** The Runner waits again only while a background
command is still running. The run ends as it does today. The reason is
in #21421.
- **A cycle that starts in the last second of a prompt turn** can have
its first messages in the prompt turn and the rest in the
provider-started turn. Tool calls stay safe: the admission rule of
#21278 refuses a call that the active turn did not announce.
- **The 5 second hold** is not tested under load. A cycle that is held
longer is dropped, and the run takes the path of today.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`) in Claude Code planned the
change and the split, reviewed the diff, and ran a part of the
verification again.
- Claude Opus 5.5 (`claude-opus-5-5`), as a Claude Code subagent with
tool use (shell, file edits, local test runs), wrote the change and the
tests and cut this slice from the full change.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-13 19:50:22 -08:00
DottaandClaude Opus 5.5 87d363a5c0 perf(ui): reveal a returning Mine visit from the saved cache (#20594)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The Mine inbox is the board's main list of work for one person; the
app saves its query cache in `localStorage` so a returning visit can
paint from it
> - On a hard load the saved cache gave no benefit: Mine restored its
rows but still held them until the first approvals, join-request and
dismissal reads answered, because those items sort in among task rows by
time and a late one would push rows down (#20407)
> - Those three reads were not saved, so every returning visit waited on
the network for items it showed a minute before
> - This pull request saves what Mine showed from those reads, as one
record per user and company, and lets it answer the wait on a returning
visit while every read refreshes in the background
> - The benefit is that a returning visit shows its rows with no API
response at all; the first visit keeps its bounded wait unchanged

## Linked Issues or Issue Description

**What happened?**
A returning hard load of `/:companyPrefix/issues?view=mine` with the
saved cache present (`paperclip:query-cache:v1:<user>:<company>`) showed
rows no sooner than with the cache cleared. On the hosted staging
instance, rows appeared about 40 ms after `GET
/companies/:id/join-requests?status=pending_approval` returned in 7 of
10 runs.

**Steps to reproduce**
1. Open Mine once so the app saves its query cache.
2. Delay API responses (for example 300 ms) and reload.
3. Rows appear only after the approvals and join-request reads answer.

**Expected behavior**
Rows show from the saved cache at once. The reads refresh in the
background.

Related: #20407 (kept approvals and join requests in the reveal gate),
#20408 and #20419 (Mine snapshot and its perf guard), #20591 (removes
the second wave of reads; merged), #20592 (task chat preload; merged).

## What Changed

- `ui/src/lib/inbox-mine-reveal-sources.ts` (new): the record `["inbox",
companyId, "mine-reveal-sources", userId]` holds the approvals visible
in Mine, the pending join requests, and the dismissals of those items
and of Mine's failed runs. A record over 256 KB is replaced by an empty
`incomplete` record that never answers the wait. `readMineRevealSources`
accepts a record only for the same user and company.
- `ui/src/lib/persisted-query-cache.ts`: the saved cache keeps that
record per user and company, only when its data names the same user and
company as its key.
- `ui/src/pages/Inbox.tsx`:
- Mine writes the record once approvals, join requests and dismissals
all have server answers.
- On a returning visit with a restored Mine snapshot, the saved
approvals, join requests and dismissals stand in while their reads are
pending, and after a transient failure. A refusal (401/403/404) or a
server answer replaces them. The first-attempt wait is skipped only when
saved values stand in; the first visit is unchanged.
- Selection fix: the selected row's key is recorded only when the
selection moves. Before, every list change recorded the row now at the
old index before the re-anchor ran, so a row inserted above the
selection (a late approval, or any live insert) moved the selection to
another row.
- `ui/src/hooks/useInboxBadge.ts`: `useInboxDismissals` takes an
optional saved `fallback` and returns `hasServerData`. Other callers are
unchanged.
- Tests:
- `Inbox.test.tsx`: a returning visit reveals rows and the saved
approval while the reads are pending, then keeps selection on the same
row when a newer approval arrives above it; a transient failure keeps
the saved approval and a 403 drops it; a record stamped for another user
is ignored and the page waits as a first visit does.
- `inbox-mine-reveal-sources.test.ts`: projection, size limit,
user/company checks, persistence scope.
- `persisted-query-cache.test.ts`: the record round-trips for the same
user, restores as invalidated, and another user's record is not saved.
- `useInboxDismissals.test.tsx`: fallback while pending, after a
transient failure, dropped after a refusal, and no fallback.
- `tests/perf/collections/mine-returning-visit.spec.ts` +
`returning-visit-fixture.ts`: production-build browser checks and the
timing case; README updated.

## Verification

- `cd ui && pnpm exec vitest run src/pages src/lib src/hooks
src/components/coordinator`: 453 files, 5574 tests pass.
- Browser checks on a disposable local instance (production UI build,
`local_trusted`, embedded PostgreSQL, Chromium headless, 1440×900):
`PAPERCLIP_COLLECTION_PERF_BASE_URL=http://127.0.0.1:<port> pnpm exec
playwright test --config tests/perf/collections/playwright.config.ts
mine-returning-visit.spec.ts`
- With every read of the company held, task rows and saved approvals
show and no company read has answered. Fails on `master`, passes here.
- An approval dismissed after the cache was saved stands in, then leaves
when the reads answer. Fails on `master` (no saved approvals), passes
here.
- Another company's Mine never shows the saved company's rows. Passes on
both.
- Timing (5 fresh contexts per arm, alternating saved/cleared, every API
response +300 ms, same instance and fixture, host load average about
20). "Responses before rows" counts API responses that finished before
the first row:

| Build | Saved cache: median (range) | Responses before rows | Cleared
cache: median (range) |
  | --- | --- | --- | --- |
| `master` (`99edcd28a7`) | 1028 ms (906–3011) | 16–19 | 1502 ms
(1423–1592) |
  | This PR | 625 ms (589–672) | 0 | 1632 ms (1359–1692) |
  | This PR + #20591 | 625 ms (604–671) | 0 | 1495 ms (1398–1585) |

Saved cache: difference of medians 403 ms. The remaining 625 ms is
JavaScript download and route-chunk loading in a fresh browser context;
no API response precedes the rows. Cleared cache: no change is claimed;
the spread overlaps.
- Existing collection specs on this build: `parent-context.spec.ts`
passes; `startup-regressions.spec.ts` has 4 failures (partial
failure/context error and two 429 cases) that fail the same way on the
`master` build. `mine-workspaces.spec.ts` was not run (needs a database
URL for its fixture).
- Review follow-up (9da56b5515), each with a red/green test that fails
on the previous head:
- Once a saved record exists, each source's answer replaces its part at
once and a refusal (401/403/404) empties it, without waiting for the
other reads, so a reload never shows refused join requests or approvals.
- The selected row's key is also recorded when the recorded row has left
the list, so a new item arriving above the fallback row never becomes
the selection or the archive target.
- An optimistic dismissal and its rollback start from the saved
dismissals that stand in, not from an empty list.
After the rebase onto `origin/master`: `cd ui && pnpm exec vitest run
src/pages src/lib src/hooks src/context src/components/coordinator`
passes (a first run under host load average ~35 had 4 timeouts and two
files missing new dependencies; after `agent-preflight` the affected
files pass, 223 tests).
- `pnpm check:token-gates`: clean. `pnpm verify` selected broader
coverage (exit 2, perf fixture is a global CI input); required CI
supplies it.
- Not run: the hosted staging page, the full local Vitest suite.

## Risks

- Saved approvals can be stale until the read answers (or after a
transient failure). A stale approval shows briefly and its actions go to
the server, which answers with the current state. A 401/403/404 drops
the saved values.
- The record holds approval payloads and join requests for the signed-in
user's company in `localStorage`. It uses the saved cache's existing
projection (secret-like fields removed), the same per-user revocation on
sign-out, and a 256 KB limit.
- #20591 is merged, so this branch (rebased on it) gets one read of each
shell request. Without it, a returning visit still reconnects the live
socket when the session is verified; because rows now reveal sooner,
that reconnect refetches more reads (here: session, health, company list
and agents twice). With both, every shell read runs once.
- Relation to #20592 (task chat preload): both edit
`ui/src/pages/Inbox.tsx` in different places and merge cleanly in either
order (checked with `git merge-tree`). This PR does not preload
anything. #20592 starts its chat preload when `allLoaded` turns true;
with this PR that happens sooner on a returning visit, while the reads
still refresh in the background, so its preload overlaps that refresh.
#20592 is merged and this branch is rebased on it with no conflicts.
- The selection change affects every Inbox tab: the key is recorded when
the selection moves, not on every list change. Archive and removal still
clamp as before.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, Playwright). Standard context window,
reasoning enabled.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 21:49:17 -06:00
Devin FoleyandPaperclip b7922570e1 fix(runtime): verify GitHub bridge health before agent startup (#21537)
Verify the inactive GitHub credential endpoint from the execution target
before publishing launcher settings. If HTTP/2 fails during startup,
stop it and verify one file-bridge attempt. Stop an unverified final
bridge before returning isolated anonymous launchers.

Regression tests cover early transport loss, delayed loss reporting,
credential isolation, and failed file-bridge cleanup. The master update
keeps the current MCP approval fixture.

Verification: 96 focused tests and server typecheck pass after conflict
resolution. Full CI passes on 843cd9945e. Greptile is 5/5 with no
unresolved review threads.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-11-13 19:41:17 -08:00
Nicky LeachandClaude Opus 5.5 4a78124a3e feat(runner): wait for a provider-started Claude turn when the controller asks (#21423)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Paperclip Runner runs Claude Code through the ACPX sidecar.
> - Claude can start a shell command in the background and end its turn.
When the command finishes, Claude Code starts a cycle of its own to
continue. Today the Runner stops Claude at the end of the turn, so the
command stops and the run fails with `native_finalization_missing`.
> - A stack of pull requests lets the Runner accept that cycle as a new
turn of the same run, when a controller asks for it.
> - `paperclip-runnerd` ends a run when a turn ends. For a Claude turn
that ended with no result while a background shell is running, this
stops the shell and the run fails.
> - #21278 binds each tool call to its turn, and #21293 reports how many
background shells are running at the end of a turn. Part 4 declared the
contract.
> - This pull request makes `paperclip-runnerd` wait for a turn that
Claude starts by itself, when the controller asks for it, and admit that
turn into the same run.
> - The benefit is that the Runner has one place that decides to wait
and one place that ends a wait.

## Linked Issues or Issue Description

No public issue exists. The issue is described here.

This pull request is one part of a change that was first opened as one
large pull request (#21316). The parts, in stack order after #21311:

1. #21419: sidecar preparation (a refactor).
2. #21420: converter for raw ACP session updates.
3. #21421: cycle stream at the adapter.
4. #21422: contract for provider-started turns.
5. #21423: `paperclip-runnerd` waits for a provider-started turn.
6. #21316: the sidecar adopts a cycle.

This is part 5. Its base branch is `feat/prp-provider-turn-contract`.
After each part, a production run does the same as today, because no
controller sends the opt-in. Refs #19912 (design).

**What existing behavior does this improve?**

What `paperclip-runnerd` does after a Claude turn ends with no result
while a background shell is running. Nothing changes until a controller
sends the opt-in.

**Subsystem affected**

Paperclip Runner: the ACPX provider backend in `runner-core` (Rust).
Claude only.

**Current behavior**

- After each completed turn the Runner emits `run.terminal`. A turn with
a running background shell gets the provider restart, and the shell
stops.

**Proposed behavior**

Nothing changes unless the controller sends `providerStartedTurns: true`
in the payload of `turn.start`. No controller sends it yet.

With that field:

- The Runner keeps the provider and waits when all of these are true:
the agent is Claude; the turn completed; the run has no accepted
`paperclip_finish` or `paperclip_block` result; the terminal event
reported `backgroundTasks` with `live` above zero; each condition of
#21293 for a kept process is true; the sidecar accepted
`session.provider_turn` with `armed: true`.
- `turn.completed` then has `awaitingProviderTurn: true`, and no
`run.terminal` follows. The Runner holds that `run.terminal`.
- When the sidecar reports a turn that the provider started, the Runner
admits it only during a wait and publishes `turn.started` with `origin:
"provider"`. In each other case that report fails the session closed.
- When a wait ends with no turn active, the Runner publishes the held
`run.terminal`. This applies when the provider process exits, the
sidecar exits, the controller cancels or stops the run, the session is
closed or suspended, and after a Runner restart.
- The held `run.terminal` and the removal of the wait are saved in one
write, and the controller gets the event from the next poll. A crash
after a cancel or a stop cannot lose it, and it is never sent two times.
- A controller `turn.start` during a wait replaces the provider first,
as before, and the held event is dropped.
- A `run.attach` from a different run during a wait ends the wait when
the attach succeeds. The Runner then discards the held `run.terminal`,
because that event is of the old run and must not end the new run. This
is the same rule that a Runner restart applies. The wait is removed only
in the state that the attach saves for the new run. If the attach fails,
for example when the old session does not stop, the old run keeps the
wait and gets its held `run.terminal` one time, from the next poll or
from a cancel or a stop. A `run.attach` from the same run publishes the
held event to that run.
- The Runner waits at most 5 seconds for the sidecar to answer a request
to arm or to disarm a provider-started turn. This is the limit that turn
cancellation already uses. Before, the limit was the 30 seconds of an
ordinary request, and a cancel, a stop or a shutdown had to wait behind
it. No answer is handled as a lost sidecar: the session fails closed,
and the Runner still publishes the held `run.terminal` one time.
- The sidecar of this branch does not know `session.provider_turn`, so
the arming request fails and the Runner does not wait. The behavior
comes with part 6.

**Reason and benefit**

- The wait is one entry in the saved session state. One rule ends it at
each save, so each new way to leave the open session ends the wait with
no new code.
- A wait must not outlive the provider or the sidecar, and a cancel must
stop it.

## What Changed

- `acpx_provider_turn_wait.rs` (new, 230 lines with its unit tests):
`provider_turn_wait { turn_id, held_run_terminal }` in the saved state;
the rule that ends a wait at save; `settle_turn_events`, which decides
at one place to arm, to create the wait and to put
`awaitingProviderTurn` on `turn.completed`.
- `acpx_provider_turn.rs` (new, 119 lines): the offer of a turn at its
terminal, the arming request, and `admit_provider_turn`.
- `acpx_provider_session.rs`, `acpx_provider_backend.rs`,
`acpx_provider_state.rs`, `provider_events.rs`,
`acpx_park_preparation.rs`: the opt-in from `turn.start`; the record
that the run has a result; the provider-started turn in the state and in
the PRP events; cancel and stop with no turn; no turn-end restart and no
park during a wait.
- `acpx_sidecar_transport.rs` (11 lines): during a wait, a poll with no
event also checks that the sidecar process is alive. A child process can
keep the output stream of a dead sidecar open.
- `bin/fake-acpx-sidecar.rs` (232 lines): provider-started turns and
failures for tests.
- Tests (1,164 lines): `tests/acpx_quiet_turn_continuation.rs`,
`tests/support/acpx_quiet_turn_restart.rs`,
`src/live/runnerd-acpx-provider-turn-wait.test.ts` (the real daemon),
`src/native-session-runtime.provider-turn.test.ts` (the controller of
today).
- `runner/DURABLE_TRANSPORT.md`: the rules.

## State saved by the previous release

- One new optional field in the saved ACPX session state:
`providerTurnWait`. It is written only during a wait. In each other
state the saved file has the shape of the previous release.
- **State from the previous release, read by this release:** the field
is absent, so there is no wait. A unit test covers this.
- **State from this release, read by the previous release (a
rollback):** outside a wait, the file loads as before. During a wait,
the previous Runner refuses the file with its "ACPX provider state is
malformed" error, because its loader refuses each unknown field. Only a
run that waits at the moment of a rollback is in this case, and no run
waits before a controller sends the opt-in.
- **Runner restart during a wait:** the wait is restored. The provider
is a new process, so the wait is over, and the controller receives the
held `run.terminal` one time. A test with the real daemon covers this.
- **Controller of today or of the previous release:** `turn.start` has
no opt-in. A test runs a turn with a running shell and a turn with an
unknown background state, and compares the events, the commands after
the turn, the restart, and the park state before and after the restart.
They are the same.

## Verification

At the head of this branch, from `packages/paperclip-runner`:

- `npx tsc --noEmit -p tsconfig.json`, `npx tsc --noEmit -p
tsconfig.test.json`, `pnpm run typecheck:typescript`, `pnpm run
build:typescript`, `node --test test/acpx-sidecar-contract.test.mjs
test/protocol-contract.test.mjs`, the server check `node
scripts/verify-runner-vendor-dependencies.mjs`, `node --test
scripts/__tests__/runner-release-crate-cache.test.mjs`, `cargo fmt
--check` and `cargo build -p paperclip-runner-core --bins` (no new
warning): all passed.
- `cargo test -p paperclip-runner-core --lib`: 644 passed, 4 ignored.
All 17 `acpx_*` suites passed.
- `src/live/runnerd-acpx-provider-turn-wait.test.ts`, 5 cases with the
real `paperclip-runnerd` and the fake sidecar through PRP: a wait ends
with exactly one `run.terminal` when the sidecar exits while a child
process keeps its output open, when the provider is lost, when the
controller cancels, and after a Runner restart; with no opt-in,
`run.terminal` follows the turn and the sidecar is never armed.
- Rust tests: wait and admit; no wait with a result; no wait with no
opt-in; an active goal neither marks nor arms a wait; a sidecar that
refuses to arm; a provider-started turn that the Runner did not wait for
fails the session closed; cancel and stop during a wait; provider and
sidecar loss; restart; the saved shape in both directions.
- `src/native-session-runtime.provider-turn.test.ts`: the real
`consumeTurn` of today gets the event sequence with no opt-in and
finalizes it as `native_finalization_missing`, as today.
- Attach during a wait, in `tests/support/acpx_quiet_turn_restart.rs`: a
different run attaches after the snapshot reports `warmAttachReady:
true`; the new run gets no `run.terminal`, the saved state has no wait,
and the old session is closed. A same-run attach gives exactly one
`run.terminal`. An attach that fails when the old sidecar is dead leaves
the wait with the old run, and the old run then gets exactly one
`run.terminal` from a poll, `run.cancel`, `turn.interrupt` or
`turn.stop`. Mutation checks: with no fix, the old `run.terminal` is in
the pending events of the new run; with the wait ended before the attach
is complete, the old run gets no `run.terminal` after a failed attach.
- No answer from the sidecar, in
`tests/support/acpx_quiet_turn_restart.rs`: the fake sidecar does not
answer the disarm request; `run.cancel` and `turn.interrupt` each return
`already_settled` in less than 10 seconds, exactly one `run.terminal`
reaches the run, `session.close` completes in less than 2 seconds with
no second `run.terminal`, and the provider process is gone. With no
answer to the arm request, the turn ends with its `run.terminal` and
with no wait. With the limit removed, each test fails at 30 seconds.
- Mutation checks, each made the named tests fail: the rule that ends a
wait at save; no `run.terminal` after a restart; arming with no opt-in;
the marker set with no wait; the sidecar process check removed.

Not run: a rollback with the binary of the previous release; a run with
the opt-in and a sidecar that answers the arming request with an error
(the code reads each error as not armed); Linux; the full suite. CI
supplies the broad coverage.

## Risks

- **The feature is off.** No controller sends the opt-in.
- **A rollback of the Runner during a wait** makes the previous Runner
refuse that saved session. See "State saved by the previous release".
- **The wait has no time limit in the Runner.** The controller must
close, suspend or cancel. Core will own that limit.
- **Park is refused during a wait.** The controller must end the wait
before it parks.
- **A new run can attach during a wait.** The attach is not blocked,
because the Runner has no time limit on the wait and a new run can send
only `run.attach`. After an attach that succeeds, the old run gets no
`run.terminal` from the Runner; the controller must not depend on that
event. The request that disarms a sidecar that the attach keeps has no
test, because no test can keep a session across an attach during a wait.
- **After a sidecar that does not answer, `runner.suspend` returns an
error at once** ("provider lifetime cleanup is not yet proven"), and
`session.close` completes. This is the behavior of today for each lost
sidecar; this PR does not change it.
- **The shared code that changed** runs for each Claude turn: the record
of a run result, and the settle step at each turn end.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`) in Claude Code planned the
change and the split, reviewed the diff, and ran a part of the
verification again.
- Claude Opus 5.5 (`claude-opus-5-5`), as a Claude Code subagent with
tool use (shell, file edits, local test runs), wrote the change and the
tests and cut this slice from the full change.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
canary/v2026.1114.0-canary.1
2026-11-13 19:21:10 -08:00
DottaandClaude Fable 5.2 689f3b190e fix(recovery): pause automatic recovery runs while a company's runs fail before starting (#21443)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Automatic recovery continues interrupted and failed tasks without a
person, with a per-task limit of 6 automatic attempts per hour
> - There is no company-level limit. When the sandbox provider refuses
to create sandboxes, every task fails the same way, and recovery keeps
creating new runs for each task
> - On a hosted instance, one hour of a provider outage made 510
automatic runs, 587 failures and 0 successes across 289 tasks. No task
had more than 4 runs, so the per-task limit did not engage. Each start
asked a refusing provider for a sandbox and used server memory, which
made the outage longer
> - Recovery must still continue every task by itself. The answer is to
wait while the infrastructure fails, not to give the task to a person
> - This pull request adds a company-scoped pause for automatic work: it
opens when recent automatic runs fail before the agent starts, the
recovery passes stop creating new automatic runs while it is open, and
it closes by itself when runs start normally again
> - The benefit is that an infrastructure outage no longer multiplies
itself through recovery, and recovery resumes with no person acting

## Linked Issues or Issue Description

**What happened?**
During a sandbox provider outage, automatic recovery created hundreds of
new runs per hour across hundreds of tasks. All of them failed before
the agent did any work. The per-task limit did not engage because no
single task reached it.

**Expected behavior**
When most recent automatic runs of a company fail before they start,
recovery waits instead of creating more runs. It tries again at a small
rate. It resumes at full rate when runs start normally again. A person's
work is never held.

**Steps to reproduce**
1. Make the sandbox provider refuse lease creation (HTTP 503) for a
company with many assigned tasks.
2. Let recovery continue the failed tasks.
3. Count automatic runs created per hour: it grows with the number of
tasks, and every run fails at sandbox creation.

Refs #21394 (continuation classification this change will reuse when it
lands), #21433 (start-pass rewrite that the start hook will build on).

## What Changed

- New table `automatic_run_brakes` (one row per company, cascade delete
with the company; migration `0480`). Migration `0481` builds
`heartbeat_runs_company_finished_idx (company_id, finished_at desc)
where finished_at is not null` concurrently, so the signal reads the
runs that finished in the window directly. It stores the state (closed,
open, half_open), the reason with the window numbers, and the
compare-and-set fields that two servers share. A company with no row is
closed.
- New module `server/src/services/recovery/automatic-run-brake.ts`:
- Signal: on the database clock, the automatic runs of the company that
finished in the last 10 minutes (and since the last state change),
newest 500 by finish time, plus up to 200 running runs. A run counts as
a failure only when it ended in its own start phase: no provider work
(`executionRecovery.providerWorkStarted = false` or the native
`nativePreDispatchFailure` record), a legacy run still preparing, or a
transient setup code. A run interrupted by a server stop, restart or
lease loss is neutral, so a release swap does not open the pause. A run
that succeeded, or ended for a person-owned or task reason, counts as
healthy.
- Opens at 12 or more counted runs with 80% or more failures, or on 5
sandbox refusals (`environment_provider_unavailable`) in a row.
- Each outcome counts once, at its own time: a finished run at its
finish time, a running run that got past setup at its start time. Read
time never orders evidence.
- Closes in two steps: open moves to half_open after at least 2 minutes
when the newest 2 counted runs are healthy or the newest one succeeded;
half_open releases at a doubling rate per minute (2, 4, 8, 16, 32, 64)
and then closes. It reopens from half_open if the signal trips again. It
closes by itself after 15 minutes with no automatic activity.
- Reads are one indexed query per company, cached for 5 seconds in a
bounded in-process map (256 companies). Each state change writes one
structured log line
(`automatic_run_brake.opened|half_open|closed|reopened`) with the
numbers.
- It also contains the start gate `holdAutomaticRunStart` (one probe at
a time, at most one new probe per 60 s, release budget per window by
compare-and-set, no I/O while the cached "held until" time has not
passed, person origin taken as an argument with a cached fallback). This
pull request does not call it from the start pass; see Risks.
- Recovery passes defer creating runs while the pause is open. The gate
is only where a run is created:
- `deliverReconciledExecutions`: after the agent-room check and right
before the wake, the delivery asks for a creation slot; when the pause
holds it, the continuation stays pending. The held backlog
(`continueHeldExecutions`) is not gated: recording a pending
continuation creates no run and costs nothing, and its delivery passes
this gate once.
- Assigned-work sweep `act()`: after the process rate limit and before
it claims a continuation, it asks for a creation slot; when held, it
returns `deferred_rate_limit` with the next try time, so nothing is
written to the per-task attempt ledger. New assignments with no earlier
run are not deferred.
- A reserved slot is given back (compare-and-set on its own timestamp)
when no run is created: the wake is refused or throws, the task is
capped, the claim changed, or the task went back to review. The
give-back is conditional in SQL on no run for that task having been
created since the reservation, so a wake that commits its run and then
throws keeps the slot.
- The sweep's overdue saved-work recovery path returns before the slot
is asked for: it starts a recovery attempt, not a run, and is never
held.
- Person work is never deferred. The gate is called only at the delivery
(`agentHasRoom`) and the sweep (`act()`), and each call passes the task
and the run it continues. While the pause is open, the gate does not
hold the record when any of these is true; it holds only when none is.
With no task or no source run it does not hold (fails open).

| Shape | How it is recognised |
| --- | --- |
| A Retry or "move forward" waiting for the task
(`deferred_issue_execution`, `queued` or `claimed`, no run yet) | an
unstarted wake request from a user for the task |
| A Retry folded (`coalesced`) into the continuation, or a Retry while
the delivery is pending | a wake request from a user for the task,
requested after the source run finished (any status) |
| A person's message or a new assignment by a person after the last run
| the same: a wake request from a user for the task after the source run
finished |
| A held task selected because a person pressed Retry (the refused Retry
run) | a run for the task created after the source run finished whose
wake is a person reason (`task_retry`, chat message), or that has
`taskRetryRequestId` or `explicitUserContinuation` |
| A continuation of a person's Retry (any depth) | the source run's
chain reaches a person's run (`listPersonOriginSourceRunIds`, the
dispatch rule) |
| A Retry during pending delivery that becomes the continuation itself
(`deliverContinuationAsRun`) | not gated: that run is created by the
person's own wake, never by these passes |
| The held backlog's saved-input path | not gated: recording or
releasing a held continuation creates no run here |
- To prevent a deadlock, a pass may still create one automatic run per
minute per company when no automatic run of the company started in the
last 2 minutes, however many runs wait (compare-and-set, shared by
servers).
- `GET /api/companies/:id/assigned-work-invariant` now returns
`automaticWork` with the state, the time it paused, the window numbers
and a plain sentence, for example "Automatic work is paused for this
company since T because 47 of 50 recent automatic runs failed before
starting."
- `listPersonOriginSourceRunIds` in `heartbeat/queued-run-line.ts` is
now exported, unchanged, so the gate uses the same person-origin rule as
dispatch.
- The outcome rules (person-owned codes, restart codes, transient setup
codes, native pre-dispatch record) come from their owning modules
through one memoised dynamic import. A static import of
`heartbeat/retries.ts` made a module cycle at server start.

### What the pause consumes, and how waste is bounded

| Token | Taken | Given back or bounded |
| --- | --- | --- |
| Creation slot (one per minute per company while open) | right before
the run is created, after every other check | released only when no run
for the task was committed after the reservation |
| Probe slot (start gate) | when the start gate lets one automatic run
through | another run may take it when the reserved run has not started
within 10 seconds (claim declined, agent paused or full) or was
cancelled, deleted or finished |
| Gradual-release start (start gate) | one per run the start gate lets
through | a declined claim spends one start of that window; the window
ends within 60 seconds and the next one doubles |
| Probe success count | once per probe run, by compare-and-set on its id
| a probe run counts once; cancelled or deleted probes count as neither
success nor failure |

### When the pause opens, stays open and closes

| State | Leaves when | Bound |
| --- | --- | --- |
| closed | opens when, in the last 10 minutes since the last change, 12
or more counted automatic runs finished and 80% or more failed before
starting, or the newest 5 counted runs were sandbox refusals | none
needed |
| open | moves to half_open when it has been open 2 minutes or more and
the newest 2 counted runs are healthy, or the newest one succeeded |
evidence arrives at least once a minute: when no automatic run started
in the last 2 minutes, a pass may create one per minute |
| open | closes when no automatic run started or finished for 15 minutes
(queued runs do not count) | 15 minutes after the last activity |
| open | stays open only while new automatic runs keep failing before
starting | each new try is spaced by the existing passes and by one
creation per minute when nothing starts |
| half_open | closes after 6 release minutes (2, 4, 8, 16, 32, 64) | 6
minutes |
| half_open | reopens when the signal trips again, counted only since
half_open began | none needed |

Restart-class endings (server stop, restart, lease loss, process loss,
interrupted native session) are neutral in every state, so a release
swap does not open the pause.

## Verification

- `npx vitest run src/__tests__/automatic-run-brake.test.ts` (14 tests,
pass): classification; opens on 20 failures before start; 50 queued
automatic runs give exactly one probe and no failure or attempt; a
person's Retry, a new assignment, a run a person joined and the
continuation of a person's Retry are not held; mixed and person-owned
outcomes stay closed; old data is ignored; 40 working runs interrupted
by a restart stay closed, then the same continuations failing at sandbox
acquisition open it; 5 refusals in a row open it; failed probe waits for
the interval, two probes past setup move to half_open, the release gives
exactly 2, 4, 8, 16, 32, 64 starts per window and then closes; reopen
from half_open; probe slot frees after cancel, delete or no start; quiet
close after 15 minutes; two instances on one database agree and start
one probe; one creation per minute while open across servers; 250 held
runs over 100 passes use at most 5 queries (at most 255 with the
person-origin fallback, one per run, once); caches stay at 256 companies
and 1024 runs.
- `npx vitest run
src/__tests__/automatic-run-brake-recovery-passes.test.ts` (pass; one
test per person shape above at the delivery and the sweep, with the
pause open and no creation slot, plus a control that still holds a
continuation no person asked for): includes "no pause record: every pass
behaves as before", "the waiting run cannot start: a pass continues a
task on another agent, then the pause closes", "one held task spends one
creation slot", and "a person's Retry or message is never paused at the
delivery, the sweep or the held backlog".: sweep, delivery and held
backlog create nothing while open, write no attempts, and continue by
themselves after recovery; the sweep still continues one task a minute
when nothing automatic waits. With the three deferral lines removed, all
4 fail (red), and they pass with them (green).
- The pause reads as closed and logs once when its run rules cannot load
(test in `automatic-run-brake.test.ts`).
- `npx vitest run src/__tests__/automatic-run-brake-import.test.ts`: the
module loads after the heartbeat service, in the server's import order.
- `npx tsc --noEmit -p server/tsconfig.json`: clean.
- `pnpm verify`: see the comment below for the result. CI supplies the
full suite.

## Risks

- This pull request does not hold starts. Already queued runs and
`transient_failure_retry` runs still start while the pause is open; they
act as the natural tries. The start gate goes in as a follow-up on top
of #21433, at a point before that pass counts a claim attempt and reads
the full run row, so a held run costs no attempt and no read.
- The sweep reuses the `deferred_rate_limit` outcome for a deferral, to
keep its outcome type unchanged while #21394 edits that file.
- The automatic-run predicate is built from master exports
(`RECOVERY_WAKE_REASONS`, retry markers). It will switch to
`isAutomaticContinuationRun` when #21394 lands.
- Migration `0480` adds one small table.
- Migration `0481` (`0481_heartbeat_runs_company_finished_index.sql`)
runs `CREATE INDEX CONCURRENTLY IF NOT EXISTS` as a top-level statement:
it is listed in `MIGRATIONS_WITH_TOP_LEVEL_STATEMENTS` in
`packages/db/src/client.ts`, like 0470 and 0476, and the runner drops
and rebuilds an invalid index after an interrupted build. The index is
declared in the Drizzle schema, so `pnpm db:generate` reports no
changes, and `pnpm --filter @paperclipai/db check:migrations` passes. On
a large `heartbeat_runs` table the build is online (no write lock) but
takes time proportional to the table. After it exists, it is one more
index on a hot table: each run that finishes writes one more index
entry.
- No user-visible text changes outside the operator report.

## Model Used

- Claude Fable 5.2 (Anthropic), via Claude Code with tool use (shell,
file edits, tests).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-13 21:20:09 -06:00
DottaandClaude Fable 5.2 6826800425 feat(ci): report the required checks on merge-queue commits (#21545)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Each merge to `master` publishes the standard image, and the time
from merge to published image limits how fast a fix reaches users
> - #21096 lets `docker.yml` push an image that the merge queue already
built, but the merge queue is not on
> - The `master` ruleset requires `ci / verify` and `ci / e2e`. Only
`pr.yml` reports them, and only for pull requests
> - A merge queue waits for the same two names on its merge-group
commit. No workflow reports them there, so every queued merge would
stall
> - This pull request adds a workflow that reports both names on
merge-group commits
> - The benefit is that the owner can turn the merge queue on with one
ruleset change

## Linked Issues or Issue Description

Refs #21096

**What existing behavior does this improve?**

The merge-queue image build from #21096. It cannot run until the merge
queue is on, and the merge queue cannot be turned on safely.

**Current behavior**

`ci / verify` and `ci / e2e` come from `pr.yml`, which runs on
`pull_request` only. With a merge queue, GitHub waits for both checks on
the merge-group commit. They never arrive. Each queued pull request
would time out and leave the queue.

**Proposed behavior**

`merge-queue-checks.yml` runs on `merge_group` and reports `ci / verify`
and `ci / e2e`. It runs no tests. It does not check out the queued
commit. It has no token permissions and no secrets.

**Reason and benefit**

A pull request enters the queue only after both checks pass on the pull
request. A queued merge therefore has the same protection as a direct
merge today. The exact-source gate still tests each `master` commit
before publication. Real tests on the combined tree can replace these
jobs later under the same names.

## What Changed

- Added `.github/workflows/merge-queue-checks.yml`. It reports `ci /
verify` and `ci / e2e` on merge-group commits.
- Added `scripts/__tests__/merge-queue-checks-workflow.test.mjs`. It
fails if the names differ from the names that `pr.yml` reports, or if
the workflow gains a checkout, an action, a secret, a token permission
or a self-hosted runner.
- `pr-trusted.yml` now runs that test and
`scripts/merge-group-image.test.mjs` in a step that skips test files an
older checkout does not have. The second test came with #21096 but no CI
step ran it.
- `scripts/merge-group-image.mjs find` now waits up to 15 minutes for a
merge-queue image build that is still queued or running. The queue can
land a commit before its image build ends. Without the wait, the
`master` push always builds from source.
- `doc/RELEASING.md` explains the workflow and lists the ruleset
settings that turn the queue on and off.

## Verification

- `node --test scripts/__tests__/merge-queue-checks-workflow.test.mjs`:
2 pass.
- `node --test scripts/merge-group-image.test.mjs`: 10 pass, with new
cases for unfinished builds.
- `node --test scripts/__tests__/release-verify-workflow.test.mjs`:
pass.
- `actionlint` on both changed workflows: no findings.
- Not verified: no `merge_group` event can fire before the queue is on.
The first queued merge is the live test. `pr.yml` calls
`pr-trusted.yml@master`, so the new CI step runs only after this merges.

Owner steps after merge:

1. In the `master` ruleset, add **Require merge queue**: Squash, build
concurrency 5, group size 1 to 5, status check timeout 60 minutes.
2. After the first queued merge, confirm that the `master` commit SHA
equals the SHA of its merge-group runs.

## Risks

- Low risk while the queue is off. The workflow has no trigger other
than `merge_group`.
- With the queue on, the queue does not test the combined tree. Two pull
requests that pass alone can still break `master` together. This is the
same as today.
- A merge-queue image build that hangs delays the source build by up to
15 minutes. A build that fails does not.
- If GitHub does not accept a job named `ci / verify` for the required
context, queued merges time out. Rollback: remove the merge queue rule.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`), Claude Code, with tool use and
code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-13 21:18:13 -06:00
DottaandClaude Opus 5.5 f6c5986b61 feat(connections): machine sign-in (MCP OAuth client credentials) for organization connections (#21473)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents use apps through connections to remote MCP servers
> - Today an organization connection must sign in as a person (browser
sign-in) or use a static key
> - Service accounts and CI-style integrations have no person who can
sign in
> - The MCP OAuth Client Credentials extension (2026-07-28) defines
machine-to-machine sign-in for this case
> - This pull request adds "Machine sign-in" for organization
connections, with a Paperclip-held signing key (recommended) or a client
secret
> - The benefit is that organizations can connect automation-owned MCP
servers safely, with credentials only in the vault and bound to one
sign-in service

## Linked Issues or Issue Description

**Subsystem affected**
Connections: remote MCP OAuth (server `tool-access/oauth`, tool gateway
token refresh) and the custom MCP server setup flow in the UI.

**Problem or motivation**
Organization-owned automation (service accounts, CI integrations) cannot
connect an MCP server that requires OAuth, because no person can
complete a browser sign-in. The MCP OAuth Client Credentials extension
(`io.modelcontextprotocol/oauth-client-credentials`) solves this with
`grant_type=client_credentials` and RFC 7523 signed client assertions.

**Proposed solution**
Add machine sign-in for organization connections only. Paperclip creates
a keypair per connection, keeps the private key in the company vault,
and shows the public key (PEM and JWK) for the admin to register. A
client secret is also supported. The sign-in service (issuer and token
endpoint) is pinned at setup. Tokens use the existing organization grant
refresh path, so caching, concurrent dedupe and 401 re-request reuse the
existing lease and single flight.

**Alternatives considered**
Reuse the deployment-wide `PAPERCLIP_TOOL_OAUTH_*` client env path that
already sent `client_credentials`. Rejected: it is not per connection,
it keeps secrets in env, and it has no issuer pin. Route machine sign-in
through `connectGalleryApp`. Rejected: its rollback deletes the draft on
a failed test, which would lose a key the admin already registered.

**Roadmap alignment**
Phase 1 of the enterprise MCP plan (no enterprise single sign-on
needed). OIDC login and enterprise-managed authorization (ID-JAG) are
later phases.

## What Changed

- `oauth/client-credentials.ts`: keypair generation (RS256, or ES256
when the service requires it), RFC 7638 key IDs, and RFC 7523 assertions
(`iss` = `sub` = client ID, `aud` = token endpoint, 60 s `exp`, unique
`jti`) with Node `crypto` only.
- `oauth/client.ts`: the token exchange accepts a signed assertion and
then sends only `client_assertion_type` and `client_assertion` (no
`client_id` or secret), as the extension specifies.
- `oauth/client-credentials-service.ts`: discovery of the sign-in
service (protected-resource metadata, then RFC 8414/OIDC metadata with
issuer check, or a token URL entered directly), issuer pinning,
vault-only key and secret handling, draft creation, test-and-activate,
and no-downtime rotation (test the new secret first; stage a new key and
switch only after the provider accepts it).
- `oauth/grant-refresh.ts` and `oauth/connection-refresh.ts`: machine
sign-in requests a new access token with the connection's own
credentials instead of needing a refresh token. It never falls back to
deployment-wide OAuth clients.
- `catalog/discovery.ts`: a rejected token on a machine sign-in
connection never re-runs discovery or rewrites the pinned endpoints.
- `mcp-http.ts` and the tool gateway: declare
`io.modelcontextprotocol/oauth-client-credentials` in client
capabilities only for connections that use it.
- Routes (with OpenAPI and activity logging): `POST
/companies/{companyId}/tools/machine-sign-in`, `GET|POST
/tool-connections/{id}/machine-sign-in[/test|/rotate|/rotate/complete]`.
Removal activity now records `machineSignIn: true`. The runner API
treats these routes as sensitive.
- Shared validators and types for machine sign-in.
- UI: "Machine sign-in" under Advanced authentication in the custom MCP
server flow (organization connections only), `MachineSignInSetup`
(choose Paperclip key or client secret, register the public key, test),
and Storybook stories for each setup state.
- Docs: `doc/connections/MCP-CLIENT-CREDENTIALS.md`, linked from
`doc/connections/README.md` and `GENERIC-REMOTE-MCP.md`.

## Verification

- `server/src/__tests__/tool-access-client-credentials.test.ts`: 16/16
pass. It uses a fake sign-in service that verifies real signatures
against registered keys, and a fake MCP server. It covers RS256 and
ES256 assertion claims and signatures, assertion-only client
authentication, the secret flow (`client_secret_basic`), discovery and
direct token URL, issuer pinning (refuses a different sign-in service
and sends it nothing), the pinned endpoint surviving a hostile
challenge, the token cache (0 token requests while live), concurrent
dedupe (5 callers on 2 service instances produce 1 token request), 401
re-request, key and secret rotation, a refused personal connection,
secret redaction in responses, rows and activity, and vault cleanup on
removal.
- Red/green: with the grant-refresh machine sign-in branch and the
discovery pin guard disabled, 5 runtime tests fail (pin, cache/dedupe,
401 re-request, both rotations). Restored: 16/16 pass. The other tests
cover new routes and services that do not exist on `master`.
- `ui/src/features/connections/MachineSignInSetup.test.tsx`: 3/3.
`ConnectionSetupFlow.architecture` and `.oauth` tests: 19/19.
- `openapi-routes.test.ts`: 51/51. `runner-api.test.ts`: 2153/2153.
`mcp-http.test.ts`: 9/9. Existing `tool-access-service.test.ts` refresh
and legacy client-credentials tests: 54/54 (filtered).
- Server package `tsc --noEmit`: clean. UI `tsc`: clean except the
pre-existing generated Storybook catalog imports.
- Not run locally (machine under heavy load): full suites, repo-wide
typecheck, browser check of the live app. CI covers breadth.

## Risks

- No migration. New config lives in `config.oauth` (`grantType:
"client_credentials"` plus `clientAuth`). Older env-configured
client-credentials connections have no `clientAuth` and keep their
original path.
- The capability declaration adds one extension key only for machine
sign-in connections. Other requests are unchanged.
- Rotation has no settings-page control yet; it is available through the
API. This is a follow-up.
- Low risk to agent boot and time to first token: a live token is read
exactly like a browser sign-in token.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, test runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 03:10:32 +00:00
DottaandClaude Opus 5.5 6b631dd477 fix(recovery): check held retained orphans only for candidate leases (#21542)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The ownership reconciliation pass lists in-progress records of each
kind every 30 seconds, and settles the records whose owner is gone
> - One kind, the retained orphan sandbox, lists leases with a large SQL
filter that has correlated subqueries on `heartbeat_runs`, `issues`,
`issue_recovery_actions` and `environment_leases`
> - The planner ran those subqueries for every released, failed and
expired lease in history, before the cheap orphan predicates. The cost
grew with lease history, not with the number of orphans
> - On a busy hosted database one page of this listing ran for 17 to 138
seconds as one statement. It held an AccessShareLock on `heartbeat_runs`
for all of that time, so an online `ALTER TABLE heartbeat_runs ADD
COLUMN` with a 5 second lock timeout failed on every try
> - This pull request runs a cheap candidate filter first and evaluates
the held checks only for candidate rows
> - The benefit is that the listing cost now follows the number of
candidates. Schema changes on `heartbeat_runs` no longer wait behind it,
and the pass reads far fewer pages

## Linked Issues or Issue Description

No public issue exists. This describes the bug in the format of the bug
report template.

**What happened?**
A hosted instance could not apply a pending migration that adds a column
to `heartbeat_runs`. The deployer sets `lock_timeout = 5s` and tries
three times. All three tries timed out. The database showed one session
that ran `select "id", "company_id", "heartbeat_run_id", "issue_id",
"metadata", "updated_at" from "environment_leases" ...` in state
`active` for up to 138 seconds. Its transaction start and query start
differed by less than 0.1 seconds, so it was one statement. It held an
AccessShareLock on `heartbeat_runs`. That instance had about 18,600
lease rows: 13,240 expired, 4,695 released, 610 failed, 69 retained, 17
active and 9 pending cleanup. Each failed try also stalled all
statements on `heartbeat_runs` for up to 5 seconds.

**Expected behavior**
The ownership listing must finish in a short time that does not grow
with lease history. It must not keep a lock on `heartbeat_runs` long
enough to block an online schema change.

**Steps to reproduce**
1. Use a database with many terminal environment leases (expired,
released or failed) and few or no retained orphan sandboxes.
2. Run the `environment_retained_orphan_sandbox` listing
(`OWNABLE_KINDS`, `listInProgress`), or wait for the ownership
reconciliation pass.
3. Run `EXPLAIN (ANALYZE, BUFFERS)` on the listing. The `heartbeat_runs`
primary key subplans run once for each terminal lease, not only for
candidates.
4. While the listing runs, run `ALTER TABLE heartbeat_runs ADD COLUMN
probe jsonb` with `lock_timeout = '5s'`. The ALTER waits until the
listing ends.

**Paperclip version or commit**
`master` at 0a96f972e1.

**Database mode**
Hosted PostgreSQL. Reproduced on embedded PostgreSQL 18.

**Additional context**
The 138 second session was not a transaction held open across remote
work. The call path is `registerOwnershipReconciliationWork` →
`runOwnershipReconciliationPass(db)` → `kind.listInProgress(db, 200,
offset)` on the pool, with no transaction. Only the `leaseKind` and
`operationKind` listings send this exact unqualified column list. Of
these, only this kind's filter reads `heartbeat_runs`.

## What Changed

- `releasableRetainedOrphanCondition` and `nextRetainedOrphanDueAt`
first apply `retainedOrphanCandidateCondition`. This is a filter with no
subqueries and no NULL results: cleanup status, run, provider, lease
status, lease policy, and the metadata keys of the three retained orphan
shapes. Expired, active and pending-cleanup leases fail it at once.
- The held checks (always held and softly held) now run inside `CASE
WHEN <retained orphan condition> THEN ... ELSE false END`. PostgreSQL
evaluates a CASE in order, so the correlated subqueries run only for
candidates. The planner can no longer split the `NOT (a OR b OR ...)`
into separate top-level subplans and run them before the candidate
predicates.
- The two `heartbeat_runs` checks of the always-held part are now one
scalar subquery with a primary key lookup. It reads
`runner_profile_json->'nativeWorkspaceSync'` once through a fenced
lateral, instead of about 12 `#>>` path reads. A scalar subquery cannot
become a hashed subplan, so the planner cannot replace it with a full
scan of `heartbeat_runs`.
- The listed rows, the order and the paging do not change.
`softlyHeldSql` is now exported so that the test can rebuild the
previous condition.
- The paging stays OFFSET-based. The `listInProgress(db, limit, offset)`
contract is shared by all ownership kinds and by the I2 rule, which
compares consecutive passes. Keyset paging is a change to that shared
contract, so this pull request does not make it.

## Verification

- `cd server && npx vitest run
src/__tests__/environment-retained-orphan-listing-cost.test.ts
src/__tests__/environment-retained-orphan-sandbox.test.ts
src/__tests__/abandoned-workspace-files-hold.test.ts`: 3 files, 55 tests
pass.
- New test `environment-retained-orphan-listing-cost.test.ts`:
- Equivalence: four seeded random fixtures of 600 leases each, across
two companies. The fixtures cover every lease status and policy, runs of
each status and runtime mode, valid and invalid native workspace
descriptors, every always-held metadata key, sibling leases on the same
sandbox, open and resolved recovery actions, open runs on the issue, and
past, future and absent expiry. Every page (limit 7 and 200) equals the
previous SQL, which is kept only in the test. `nextRetainedOrphanDueAt`
also equals the previous query.
- Red/green: over 2,000 history leases with the hosted status mix (71%
expired, 25% released, 4% failed), the test counts the `heartbeat_runs`
rows that the listing reads, from `EXPLAIN (ANALYZE, FORMAT JSON)`. On
`master` it reads 1,760 rows for 1 candidate, so the test fails. With
this change the count is within the bound of 10 per candidate.
- Incident regression: while listings run in a loop over 5,000 history
leases, `ALTER TABLE heartbeat_runs ADD COLUMN ...` and `DROP COLUMN`
with `lock_timeout = '5s'` complete in less than 5 seconds.
- Local measurement (embedded PostgreSQL 18, Apple M-series, warm
cache), not in CI:
- 18,640 leases with the hosted status mix and a 19 KB runner profile
per run. `master`: 422 ms, 296,771 buffers (12,357 read). This change:
26 ms, 36,004 buffers. The `heartbeat_runs` subplans run 13 times (once
for each candidate), not once for each lease.
- 300,000 history leases. `master`: one listing page takes 2,822 ms, and
the ALTER under load waits 2,448 ms. This change: 212 ms, and the ALTER
takes 96 ms.
- `pnpm check:module-boundaries` passes.

## Risks

- Low risk. The listing and the due-time query return the same rows, and
the equivalence test compares them. The settle is unchanged and still
re-checks every hold under the row lock.
- The candidate filter must stay a superset of
`retainedOrphanSandboxCondition`. If a later change adds a retained
orphan shape and does not update the candidate filter, the listing skips
that shape. The equivalence test catches this only for the shapes in the
fixture.
- The listing still reads each lease row once with cheap predicates. A
partial index on the candidate filter could make it scale only with
candidates. This pull request does not add a migration.
- The ownership pass still lists all kinds every 30 seconds when its
fingerprint changes. This pull request does not change that cadence.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with tool use
(shell, file edit, embedded PostgreSQL runs).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:48:21 -06:00
DottaandClaude Opus 5.5 1dee321844 perf(run-events): read the wake's transaction id and the replay probe in the lock read (#21528)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Every native event that announces work to another part of the server
(chat progress, a final response, an evidence outcome, recovery)
registers a wake in its transaction before it writes the event
> - Such an event takes six statements: BEGIN, the run lock read, a
replay probe, a read of the transaction id for the wake, the append, and
COMMIT. A controller checkpoint save takes five: BEGIN, the owner lock,
the transaction id, the write, COMMIT
> - In the agent scaling load model these two paths are now the largest
per-agent cost after #21511 and #21520, and other events fall behind
first at 600 agents
> - This pull request returns the transaction id from the lock read that
each path already sends, and moves the replay probe into the run lock
read when it is safe
> - The benefit is four statements for such an event and four for a
checkpoint save, with the same locks, the same wake order and the same
replay rule

## Linked Issues or Issue Description

Refs #21511 and #21520 (the delta batch path). Refs #21418 (agent
scaling plan and load model).

**What existing behavior does this improve?**
The statement count of native events that register a wake
(`appendNativeSourceEvent`) and of controller checkpoint saves
(`nativeRunnerCheckpointWriter`).

**Subsystem affected**
`packages/db/src/work-signals.ts` (one new export),
`server/src/services/heartbeat-run-events.ts`,
`server/src/services/native-runtime/native-runner-checkpoint.ts`.

**Current behavior**
`signalDatabaseWork` sends `select pg_current_xact_id()` for the first
wake of each transaction. A native event with a wake reads the run lock,
then probes for its replay in a separate statement, because a statement
that waits for the lock keeps its earlier snapshot.

**Proposed behavior**
- The run lock read and the checkpoint owner lock read also return
`pg_current_xact_id()::text`. Both already lock a row, so the
transaction has an id at that point. `primeDatabaseWorkTransactionId`
gives it to the transaction's wake state, and the wake sends no
statement for it.
- For an event that may register a progress, response, reassessment or
retention wake, the run lock read also reads `next_event_seq` from its
own snapshot and checks for a replay in that snapshot. Every event
append advances `next_event_seq` in its transaction. So when the locked
row has the same `next_event_seq` as the snapshot, no append committed
in between, and the probe result is current. Otherwise, or when the
probe finds a replay, the separate probe runs as before.

**Reason and benefit**
Fewer round trips while a pool connection is held in a transaction. In
the load model these paths hold the main pool and the three-connection
checkpoint pool longest per event.

**Breaking changes**
None. No schema change.

## What Changed

- `primeDatabaseWorkTransactionId(connection, xid)` (`work-signals.ts`,
exported from `@paperclipai/db`): inside a transaction whose id is not
yet known, it records the id. Outside a transaction it does nothing. An
uncertain commit still resolves with `pg_xact_status` on the same id.
- `appendNativeSourceEvent`: the lock read returns the transaction id
when the event has any wake. For a probing wake it is one statement: a
CTE that reads `next_event_seq` from the snapshot, a CTE that locks the
run row (`for no key update`, unchanged), and `exists (...)` for the
replay. The wake is still registered after the replay check and before
the append.
- `lockNativeRunnerCheckpointOwner` returns the transaction id with its
owner lock and primes the wake state. Its return value and every caller
are unchanged.

## Verification

- Red/green on embedded PostgreSQL; each new assertion failed against
`origin/master` first:
- `heartbeat-run-event-atomic-append.test.ts` "registers a native
event's wakes before its write and only when the event is new": four
statements (before: six), the wake's intent and settlement are
published, and a replay publishes no wake.
- New: "finds a replay that another writer committed while the lock read
waited, and registers no wake for it". Another connection holds the run
row (the test waits until PostgreSQL reports the append waiting for that
lock), stores the same event and moves `next_event_seq`; the waiting
append returns `duplicate`, runs the separate probe, and publishes no
wake. With the `next_event_seq` comparison removed, this test fails.
- New `native-runner-checkpoint-save.test.ts`: a save is BEGIN, the
owner lock with the transaction id, the write, COMMIT (before: five
statements), and its recovery wake is published; a save after the lease
expired writes nothing and fails with
`native_runner_checkpoint_owner_lost`.
- `work-signals.test.ts`: a primed transaction sends no id statement for
its wakes; a primed transaction whose connection is killed after its
write stays pending and then resolves through `reconcileDatabaseWork`,
with no row stored.
- Added after merging `origin/master` (with #21511, #21520 and #21436)
into this branch:
- "gives the wake the transaction id of the transaction that took the
run lock": inside a caller's transaction, an event with a progress wake
sends no separate `pg_current_xact_id()` statement; the transaction's
connection is then killed, and `reconcileDatabaseWork` checks
`pg_xact_status` with exactly the id that `pg_current_xact_id()` returns
on that same transaction. Priming a wrong id makes it fail.
- "primes nothing when the lock read fails, so the transaction is not
left uncertain": a lock read that fails on `lock_timeout` leaves no
pending work; after the same failure inside a savepoint, the next signal
on that transaction still sends its own `pg_current_xact_id()`. Priming
before the lock read makes it fail.
- Both call sites pass only the value returned by the statement they
just ran on the same `tx`: `appendNativeSourceEvent` (`tx.execute(...)`
then `primeDatabaseWorkTransactionId(tx, run.xid)`) and
`lockNativeRunnerCheckpointOwner` (`tx.with(locked).select(...)` then
`primeDatabaseWorkTransactionId(tx, xid)`).
- Retest on the combined tree, one file at a time: atomic-append 28
(including the equivalence test and both race tests), the port suite
with conformance 105, checkpoint-save 2, `live-recovery-work` 51,
`work-signals` 25, runner `native-session-runtime` 297; server typecheck
passes. No code change was needed for the combination.
- Other focused suites pass, one file at a time: the port suite with its
conformance suite (100), `chat-native-progress-work` (35),
`native-observer-replay` (1), `heartbeat-run-event-sequencing` (5),
`native-event-boundary-golden` (1), `native-recovered-owner-followup`
(97), `native-restart-recovery-isolation` (1),
`heartbeat-native-restart-suspension` (115), `live-recovery-work` (51),
`work-signals` (25).
- `pnpm --filter @paperclipai/server run typecheck` and `pnpm
check:module-boundaries` pass.
- Load model (`scripts/benchmark-agent-scaling-load.ts` as merged in
#21418). Stacked: this pull request merged locally with #21511, and the
model's local `DELTA_COALESCE=1` mode that mirrors #21520 (the model
applies the port's batching rules itself; that mode is not in any pull
request). Default pool of 10 connections; embedded PostgreSQL on the
same, shared machine; runs one after the other with a pause between
them. No run had a failed operation, a failed checkpoint save or a lost
lease. "Before" rows are `origin/master` from the same session.

| Agents | Code | Window | Load avg | Event-loop utilization |
Statements/s | Delta lateness p50 / p99 (ms) | Other events done /
offered | Other event lateness p50 / p99 (ms) |
  | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 200 | before | 60 s | 9.6 | 0.56 | 7,043 | 283 / 3,368 | 12,000 /
12,000 | 278 / 3,895 |
| 200 | stacked | 60 s | 7.9 | 0.31 | 2,725 | 102 / 253 | 12,000 /
12,000 | 1 / 70 |
| 400 | before | 60 s | 11.8 | 0.59 | 7,158 | 9,927 / 18,779 | 6,068 /
24,000 | 1,904 / 60,154 |
| 400 | stacked | 60 s | 8.6 | 0.60 | 5,609 | 117 / 955 | 24,000 /
24,000 | 2 / 1,274 |
| 600 | before | 60 s | 15.0 | 0.85 | 12,419 | 7,108 / 11,017 | 7,836 /
36,000 | 10,822 / 52,171 |
| 600 | #21511 + #21520 only | 120 s | 11.4 | 0.72 | 6,730 | 1,147 /
6,335 | 61,158 / 72,000 | 11,678 / 21,212 |
| 600 | stacked | 120 s | 9.6 | 0.93 | 10,290 | 656 / 4,233 | 71,895 /
72,000 | 1,136 / 4,980 |
| 600 | stacked (repeat) | 120 s | 9.8 | 0.90 | 9,579 | 648 / 4,509 |
71,958 / 72,000 | 1,319 / 5,712 |
| 800 | stacked | 60 s | 9.7 | 0.98 | 10,971 | 920 / 3,305 | 46,244 /
48,000 | 1,974 / 3,835 |

With the three changes, one process finishes 99.9% of the offered work
at 600 agents, but its event-loop utilization is 0.90–0.93, above the
0.75 target, and p99 lateness is about 5 s. It saturates between 600 and
800 agents.
- Statements: an event with a wake 6 → 4; a checkpoint save 5 → 4 (6 → 5
where commit-time notifications are on).

## Risks

- The transaction id is read in a statement that already assigns one (a
row lock), so no transaction gets an id it did not have before.
- The probe in the lock read is used only when `next_event_seq` did not
change between the snapshot and the lock. Every writer of
`heartbeat_run_events` moves `next_event_seq` in the same transaction
(native and ordinary appends, the run coordinator store, the retry
queue). A future writer that inserts events without it would make this
probe miss a concurrent replay; the append statement then still finds
the replay and stores nothing, but the wake would already be registered
(a spurious wake, not a lost or duplicate event).
- This pull request and #21511 both edit `heartbeat-run-events.ts`, in
different functions.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:48:17 -06:00
Devin FoleyandPaperclip dc018f9bc0 Keep agent-owned connections healthy during catalog refresh (#21539)
Use dedicated agent grants for catalog refresh, GitHub health checks, and reconnect status. Preserve shared connection health when a different agent lacks authorization.

Add PostgreSQL and HTTP regressions for credential selection, stale health recovery, caller isolation, and actual provider rejection. Correct the MCP demo approval fixture to submit a valid saved-card option through normal answer validation.

Validation: 629 focused tests, server typecheck, all CI checks, and Greptile 5/5 with no unresolved review comments.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-11-13 18:45:20 -08:00
DottaandClaude Opus 5.5 833689a3fc fix(queue): say a long-queued turn is waiting to start, and name the routine execution that holds it (#21447)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - #21405 made a task that waits for its agent's concurrency limit show
"Waiting for <agent> · N slots max · P of Q in line · Configure"
> - On hosted staging, after that deploy, a routine task stayed queued
for 8 minutes while its agent had 19 free slots. Its chat said "Working
for 7m · Thinking", and the server label was a plain "Queued"
> - The real cause was the routine's one-execution-at-a-time rule: the
previous execution of the same routine was still open with its run. That
is correct scheduling, but the UI did not say it
> - A queued turn also kept the startup feedback ("Working") for as long
as it stayed queued, with no limit
> - This pull request names the routine hold and limits the startup
feedback to 30 seconds
> - The benefit is that a queued task always says that it waits, and why
when the server knows

## Linked Issues or Issue Description

Refs #21405

**What happened?**

On staging, one routine task's continuation run was queued at 17:39:03Z
and claimed at 17:47:10Z. The agent ran 1 of 20 runs. The chat showed
"Working for 7m · Thinking". The run's projection label was "Queued"
with no cause.

**Why did it happen?**

1. The routine allows one open execution with a run at a time (unique
index `issues_open_routine_execution_uq`). The previous execution of the
same routine was still open and held its run until 17:47:10.172Z. The
waiting run was claimed 345 ms after that task closed. The claim was
refused with "Another execution for this routine is already in progress"
until then. This is the routine's intended policy, not a scheduling bug.
2. The execution projection did not name that hold, so the server label
was "Queued".
3. The task chat keeps the startup feedback ("Working" and "Thinking")
for a queued turn with no named hold (a DESIGN.md rule). That rule had
no time limit.

**Expected behavior**

A routine task that waits for the routine's open execution says "Waiting
for <task> to finish". A queued turn that has no named hold says
"Waiting to start" after a short grace period.

## What Changed

- Server: `executionProjectionsForRuns` reads, in one query per batch,
whether another open task of the same routine holds the routine's
execution. For a never-started queued run that waits on it, the
projection sets the cause `routine_execution_in_progress`, the label
"Waiting for <task identifier> to finish", and a next action. This check
comes before the agent-limit check, because a free slot would not start
the run.
- UI: a new hook, `useWaitingToStart`, keeps the startup feedback for 30
seconds. After that, a queued turn with no named hold shows "Waiting to
start": an hourglass, no spinner, no shimmer. The task chat's runner
header, its current-activity tail and the live run pill use it. The hook
uses one timer at the threshold and no per-second ticks.
- DESIGN.md: the startup-feedback rule now has the 30-second limit and
the routine hold.
- Storybook: `WaitingToStart` and `WaitingForRoutine` stories, with play
checks.

**Why 30 seconds:** DESIGN.md keeps "Working" from message submission
through admission, so that a normal start does not flicker. On staging,
runs that were not held started 3 to 8 seconds after they were queued
(from the company's run list between 17:30 and 17:48Z). 30 seconds is
several times that, and shorter than any real wait the user should be
told about.

## Verification

- Red/green server test (embedded Postgres),
`execution-projection-agent-queue.test.ts`: a routine task waits while
another open task of the same routine holds a run. The test checks the
label "Waiting for Q-3093 to finish" and the cause. It checks that the
holder is not marked as waiting on itself, and that the label returns to
plain "Queued" when the holder closes. It fails without the server
change.
- Red/green UI tests: `TaskChatRunnerTurn.test.tsx` (startup feedback
until the threshold, then "Waiting to start" in the header and the tail;
a running turn is never called waiting) and
`TaskChatLiveRunPill.test.tsx` (the label changes, with no spinner or
shimmer). Both fail without the UI change.
- `vitest run` on the four `execution-projection-*` server suites (36
tests) and all `task-chat` and `TaskChatThread` UI suites (1,414 tests).
UI and server typecheck.
- The new stories render in headless Chromium.
- On staging, before this change, the #21405 waiting line was confirmed
with a temporary 1-slot agent: "Waiting for Queue Check (temp) · 1 slot
max · 2 of 2 in line · Configure". The temporary tasks and agent were
cancelled and terminated.
- After deploy: confirm the new labels on staging.

## Risks

- Low risk. There is no schema change.
- One extra query per projection batch that includes a never-started
queued run on a routine task. It is a join on the company's
routine-execution tasks through the existing `issues_company_origin_idx`
index.
- A queued turn that is admitted after 30 seconds, for a reason the
server does not name, now says "Waiting to start" instead of "Working".
This is the intended change.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with tool use
(shell, file edits, Chrome for staging checks).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:22:07 -06:00
DottaandClaude Opus 5.5 a82e235828 fix(heartbeat): keep recovery behind people's work, guarantee a start to tasks people wait on, and admit recovery a few runs at a time (#21427)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The heartbeat queue starts an agent's queued runs when a slot is
free. #21234, #21277, #21349 and #21391 order them: person work first, a
recovery share, aging, and a share for the longest-waiting run
> - On hosted staging, one agent with 20 slots had 295 queued runs. 139
of them were system recovery wakes that entered the queue within one
hour. Runs that people asked for waited 7 hours
> - Two rules let this happen. Recovery took every fourth start even
when a person's run waited, and an aged recovery run moved into the
agent-work tier ahead of todo tasks. The continuation producer also
admitted any continuation held past 5 minutes, however deep the queue
was
> - This pull request keeps recovery behind people's work, gives a
person's long-waiting run a bounded start, re-checks a recovery run
before it starts, and admits recovery to an agent's queue a few runs at
a time
> - The benefit is that a task a person asked for starts within a
bounded number of dispatches, and a recovery burst cannot flood an
agent's queue

## Linked Issues or Issue Description

Refs #21234, #21277, #21349, #21391, #21405, #21433, #21502

**What happened?**

On hosted staging (2026-11-13), one agent had `maxConcurrentRuns` 20 and
all 20 slots held hour-long runs. 295 runs were queued. 139 had wake
reason `issue_recovery_action_restored` (source `execution.reconciled`),
enqueued between about 10:46Z and 11:47Z. Two tasks that people created
were stalled by a restart at 07:25Z. Their next runs were queued at
07:29Z and were still queued 7 hours later. Both were system wakes: one
`issue_monitor_due`, one `issue_continuation_needed` from the
assigned-work sweep. Recent starts had waited from 1 to 290 minutes.

**Why did it happen?**

1. The recovery share gave every fourth start to system recovery, also
while a person's run waited.
2. A recovery run that waited one hour aged into the agent-work tier.
Recovery tasks are usually in progress, so they sorted ahead of todo
tasks in that tier.
3. The oldest-run share served the oldest aged run of any class. Older
agent work and older recovery came before a person's run.
4. Only a wake that a person requested counted as person work. A system
wake on a task that a person created and waits on (a monitor check, a
continuation after a restart) competed as ordinary work or recovery.
5. `deliverReconciledExecutions` admitted a continuation held past
`RECOVERY_CONTINUATION_MAX_HOLD_MS` (5 minutes) whatever the queue
depth. At one per agent per 15-second pass, one agent got about 139
continuations in an hour.

**Expected behavior**

A run that a person created or asked for never waits behind system
recovery. Waiting never fails or cancels it. A long wait moves it
forward, so it starts within a bounded number of dispatches. Recovery
enters the queue a few runs at a time. A recovery run that is no longer
needed is dropped quietly.

**Steps to reproduce**

1. Give an agent 20 slots, all busy.
2. Queue 60 agent assignments, then two system wakes on tasks people
created (a monitor check and a continuation), then 30 later person runs
on in-progress tasks, then a burst of 139 recovery wakes.
3. Free one slot at a time while people and agents keep adding work.
4. See that the monitor check starts only at dispatch 244.

## What Changed

- `server/src/services/heartbeat/queued-run-dispatch-order.ts`:
- System recovery has its own tiers below agent work (fresh, then
backlog). Aging moves old backlog only into the fresh recovery tier,
never ahead of agent work.
- The recovery share applies only when no ready person run waits. It
comes out of agent work, never out of a person's turn. Within the share,
the longest-waiting aged recovery run goes first.
- The oldest-run share serves the longest-waiting aged run that a person
asked for or whose task a person created (`issues.created_by_user_id`),
first. This includes a system continuation of such a task. Then it
serves the longest-waiting aged agent-work run. It never serves other
recovery.
- New dispatch fact `personTask`, read with the queued tasks' status and
priority (same query).
- `queuedRunDispatchLine` (the O(n log n) place-in-line order from
#21405) follows the same rules. `queued-run-line.test.ts` still proves
it equals the dispatcher's order over 400 random queues, now with
`personTask` in them.
- `server/src/services/heartbeat/queued-run-line.ts`: the start-history
counters use `dispatchHistoryStart` and `dispatchCountersFromStarts`, so
dispatch, the place-in-line view and the tests share one rule.
- New pure helpers `oldestShareCandidate`, `dispatchHistoryStart` and
`dispatchCountersFromStarts`. The queue and the tests use the same
counter rule (which includes the #21391 rule that a person's recovery
restart does not use up the recovery share).
- `server/src/services/heartbeat/queue.ts`:
- At claim, a system recovery run (`issue_recovery_action_restored`,
`issue_continuation_needed`, not requested by a person) is re-checked.
If the task is reassigned, or another run continues it (started after it
was queued, running, a queued non-recovery run, or a newer queued
recovery run), the run is cancelled before it starts with
`recovery_no_longer_needed` and its wake is skipped. This is the same
quiet path as the existing closed-task check: no failure, no provider
work. A person's undelivered message on the run still keeps it.
- `server/src/services/recovery-queue-admission.ts` (new):
`QUEUED_RECOVERY_RUNS_PER_AGENT` = 3 and the queue-depth check.
- `server/src/services/execution-recovery-resolution.ts`: a continuation
is admitted only while its agent has fewer than 3 queued recovery runs,
fresh or backlog. The 5-minute hold bound still orders held
continuations, but no longer overrides the cap. The rest stays pending
in its recovery action. The existing continuation pass (every 15 s while
any continuation is pending) admits the next one after a queued recovery
run starts or is dropped. No new polling.
- `server/src/services/recovery/legacy-workspace-rehome.ts`: the
pending-wake delivery for rehomed and released-writer tasks uses the
same cap. A held wake waits 60 s for its next turn and does not count as
a refusal.
- The oldest-run share counts as served only by a start that people wait
on while its candidate is one, so older agent work never uses up a
person's turn (`servesOldestShare`, used by the dispatcher, the
place-in-line order and the start history, which now records
`personWaiting`).
- `server/src/services/recovery/assigned-work-sweep.ts`: a rehome or
released-writer wake held by the cap (`HELD_RECOVERY_WAKE_KEYS`) counts
as the task's pending path while its assignee has queued recovery runs.
- `server/src/services/recovery/assigned-work-sweep.ts` (with #21502's
stalled-queued-run replacement): a start of work people wait on does not
count as passing over a queued run that people do not wait on. Waiting
behind people's work is intended, so the sweep does not cancel such a
run and replace it with a recovery continuation. Test:
`assigned-work-stalled-queued-run.test.ts`, "leaves an automatic run
that waits behind work on tasks people created". It fails without the
rule.
- `server/src/services/recovery/assigned-work-sweep.ts`: a pending
continuation still counts as the task's wake path past
`PENDING_CONTINUATION_BOUND_MS` while its agent has queued recovery
runs, so the sweep does not treat a held task as stranded.

**Bounds**

- A run a person asked for, or any run on a task a person created: once
it has waited past `QUEUED_RUN_AGING_MS` (1 hour), it starts within
`OLDEST_DISPATCH_SHARE` × (such aged runs queued before it + 1)
dispatches. Before that, runs a person requested already go first. There
is no time limit, and waiting never cancels it.
- Recovery run, proposed bounded behavior: at most 3 queued per agent.
When no person run waits, an aged recovery run starts within
`RECOVERY_DISPATCH_SHARE` × (aged recovery runs before it + 1)
dispatches. At start it is re-validated against the current task, and it
is dropped quietly if stale. While people keep the agent busy, recovery
waits. It is held durably in its recovery action, not lost and not
failed.

## Verification

Regression test that reproduces today's shape
(`server/src/services/heartbeat/queued-run-dispatch-order.test.ts`,
"deep per-agent queue: a recovery burst behind 20 busy slots"). It uses
20 busy slots, one dispatch per freed slot, 60 older agent assignments,
the two 07:29 system wakes on tasks people created (monitor check and
continuation), 30 later person runs, a 139-run recovery burst, and new
person and assignment arrivals on every step. The same model, run
against master's dispatcher and this one over 600 dispatches:

| | Before (master) | After |
| --- | --- | --- |
| Dispatch at which the 07:29 monitor check starts | 244 (≈12 h at ~20
starts/h) | 4 |
| Dispatch at which the 07:29 continuation starts | 5 | 8 |
| Recovery starts while a person-requested run waited | 89 | 0 |

With person-requested runs in the 07:29 slots instead (the first version
of the test), master starts them at dispatch 244 and 248, and this
change at 4 and 8.

- Red on master: the new dispatch tests fail (`expected -1 to be greater
than or equal to 0`, and recovery started ahead of agent assignments).
Green with this change.
- Admission cap, embedded Postgres (`recovery-wake-herd.test.ts`,
"queues at most a few recovery runs per agent..."): with the cap
removed, 12 of 12 continuations enter one agent's queue in 12 passes.
With the cap, 3 enter, the rest stay pending, and all 12 are delivered
as queued runs leave. The queue never holds more than 3 recovery runs.
- Older agent work (`queued-run-dispatch-order.test.ts`): with 60 older,
higher-priority assignments and no person arrivals, an aged
`issue_monitor_due` run on a person-created task starts within
`OLDEST_DISPATCH_SHARE` in dispatch and in the place-in-line order. With
the old reset rule it does not start in 8 dispatches.
- Held restored-task wake (`legacy-workspace-rehome.test.ts`): with the
agent's recovery queue full, the wake stays pending with no failure
counted, the sweep reads the task as `pending_intent`, and the wake is
delivered after a queued recovery run leaves.
- Start-time re-check, embedded Postgres, real heartbeat queue
(`heartbeat-stale-queue-invalidation.test.ts`): a recovery run whose
task was reassigned, or that a newer run already continued, is cancelled
with `recovery_no_longer_needed`, its wake is `skipped`, and the adapter
is never called. A recovery run that the task still needs starts.
- Updated tests that encoded the old rules: recovery's share and aging
now come out of agent work, not out of a person's turn. Sweep liveness
while held is covered.
- `pnpm --filter @paperclipai/server exec tsc --noEmit`: passes.

## Risks

- Behavior change: a system run on a task a person created now takes the
oldest-run share ahead of older agent work. On an agent whose queue is
mostly such tasks, agent-made subtasks get that share less often. They
still start in normal order.
- Behavior change: while people keep an agent fully busy, system
recovery for that agent waits. This is intended. Recovery is held
durably and starts once no person run waits.
- Behavior change: a queued system recovery run can be cancelled at
start as no longer needed. It is cancelled before any provider work,
with the same quiet outcome as a continuation whose task closed.
- A continuation held behind the cap stays pending longer than 7
minutes. The sweep now treats it as live while its agent has queued
recovery runs. If such queued runs can never start (for example
dependency-blocked), the next continuations wait behind them.
- One extra indexed read per recovery claim, on the existing
`(company_id, context_snapshot->>'issueId', created_at)` index. No
schema change. Mixed versions are safe: an old server keeps its old
order until it is replaced.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, in Claude Code with
tool use (shell, file edit).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:21:35 -06:00
DottaandClaude Opus 5.5 a726dce4f0 feat(mcp): show live progress for long connected-app tool calls in the task chat (#21449)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents call tools from connected apps through the Paperclip MCP
gateway
> - Some tool calls take a long time: exports, builds, large searches
> - The MCP progress utility lets an app report how far a call is, but
Paperclip ignored it
> - So the task chat showed only a spinner until the call finished, with
no sign of life
> - This pull request reads the app's progress and shows it on the
tool's row while the call runs
> - The benefit is that people can see that a long call is working, how
far it is, and how long it has run

## Linked Issues or Issue Description

**Subsystem affected**
Connected apps (MCP gateway) and the task chat.

**Problem or motivation**
A long tool call from a connected MCP app shows only a spinner in the
task chat. Paperclip does not send a `progressToken`, and it drops
`notifications/progress` from the app.

**Proposed solution**
Send a `progressToken` on each `tools/call` that an agent run makes.
Read the app's progress notifications from the call's response stream.
Show a progress bar (or an indeterminate sweep), the latest message, and
elapsed time on the running tool row. Keep the latest state so that a
reload mid-call shows it.

**Alternatives considered**
A new transport for progress. The existing company live-events socket
already carries per-run events with run read checks, so this change uses
it.

**Additional context**
Spec:
https://modelcontextprotocol.io/specification/2026-07-28/basic/utilities/progress

## What Changed

- `mcp-http.ts`: `mcpHttpRequestInit` can add `_meta.progressToken`.
`readMcpHttpResponse` gives each notification on the response stream to
an observer before the result. The observer cannot fail or delay the
call. New `mcpProgressNotificationOf` matches a notification to its
token.
- `tool-gateway.ts`: remote HTTP and local stdio `tools/call` from an
agent run send a per-call token and forward progress to a reporter.
Calls without a run send no token.
- New `server/src/services/tool-call-progress.ts`: checks each update
(finite, increasing, plain single-line text of at most 200 characters).
It publishes a `heartbeat.run.tool_progress` live event at most every
250 ms (latest wins) and saves the latest update on the invocation at
most every 1 s. Nothing is awaited by the call. A final `done` event
goes out when the call ends.
- DB: nullable `tool_invocations.progress` jsonb column (migration
`0479`, `ADD COLUMN` only). A save never writes onto a finished call.
- Live delivery: the new event follows its run like transcript output. A
client gets it only for runs it follows, after the run read check, in
the run's company only.
- New route `GET /api/issues/:issueId/tool-call-progress` (task read
check) returns the latest progress of the task's running calls, for a
reload mid-call. Documented in `docs/api/issues.md`.
- UI: `ToolCallProgressProvider` (one per chat) keeps the latest update
per call. A running tool row subscribes and shows a bar or sweep, the
message, the percent and elapsed time (design tokens, reduced motion
honored). Finished rows drop it and never fetch or re-render for
progress. `LiveUpdatesProvider` ignores the event for query
invalidation.
- `packages/kv-demo-mcp-server`: new `slow_with_progress` demo tool
(`steps`, `stepMs`, `withTotal`) in its own file.
- Storybook: `Tasks/Tool progress` stories (running with total, running
without total, done).
- Docs: new `doc/connections/MCP-PROGRESS.md`;
`MCP-PROTOCOL-VERSIONS.md` no longer lists progress as unsupported.

## Verification

- Red/green: on the original source, 5 of the new server tests fail
(token in `_meta`, SSE with interleaved progress and result,
notification matching, run-interest delivery, gateway end-to-end). With
this change they pass.
- `server/src/__tests__/mcp-http.test.ts` (13),
`tool-call-progress.test.ts` (9: throttling, latest wins, bounded text,
out-of-order drop, finish, failures never throw, company/run/tool
scoping), `tool-call-progress-store.test.ts` (2, embedded Postgres:
company and task scoping, no write onto a finished call),
`realtime/live-event-delivery.test.ts` (run interest),
`tool-gateway.test.ts` end-to-end against the real SDK kv demo server
over Streamable HTTP: 6 provider updates over about 1.2 s give 3 to 7
throttled events, all for the right company, run, task and tool, the
last one `done`; no event reaches another company; the invocation keeps
the progress; the result is unchanged.
- UI: `tool-call-progress.test.tsx` (10: name matching, row states,
plain-text rendering, drop on complete, restore after reload, no fetch
for finished rows). All 67 task-chat test files pass.
- `packages/kv-demo-mcp-server`: 13 tests, including progress over HTTP
with and without a total.
- Storybook: checked the three stories in a browser. The bar fills to
40%, the sweep animates, the elapsed time ticks, and the done row shows
no progress.
- `pnpm verify`: see the CI checks for the final revision.

## Risks

- Low. The migration adds one nullable column.
- An SSE response is now bounded per message, not per stream, so handled
notifications (such as progress) do not use up the result's size limit.
Each message is still bounded.
- Progress code runs only while a run's tool call is in flight. It adds
a few bytes to the request and throttled, fire-and-forget work. Agent
boot and time to first token do not change.
- A row finds its call by run and tool name. If one run calls the same
tool twice at the same time, both rows show the newer call's progress.
- Progress from a call that became an MCP Task is not shown.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Claude Code agent with tool use
and code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 20:14:22 -06:00
DottaandClaude Fable 5.2 5d0b27acca fix(run-events): watch both late appends for rejection as soon as they start (#21536)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The server stores each run event through `appendHeartbeatRunEvent`
and `appendHeartbeatRunEventBatch`
> - Pull request #21511 added a test that deletes a run while two
appends wait for its row lock
> - Both appends must fail, but the test attached the rejection check of
the second append only after it awaited the first
> - When the second append failed first, its rejection had no handler,
and Vitest failed the whole shard with one unhandled error
> - This pull request attaches both rejection checks when the appends
start
> - The benefit is that server test shards on master and on unrelated
pull requests pass again

## Linked Issues or Issue Description

Refs #21511

**What happened?**

Server test shards failed on master and on unrelated pull requests.
Every test passed. Vitest reported one unhandled error: `Error:
heartbeat_run_event_binding_mismatch`, from
`server/src/__tests__/heartbeat-run-event-atomic-append.test.ts`.

**Expected behavior**

The test awaits both expected failures. No rejection stays without a
handler.

**Steps to reproduce**

Run the general server test shard that contains
`heartbeat-run-event-atomic-append.test.ts` on master at `cefef8f11d`.
Examples: master "verify / General tests (server (1/10))", pull request
#21443 "ci / General tests (server (9/12))", pull request #21505 "ci /
General tests (server (1/12))".

## What Changed

- In the test "waits for Stop on the run row and appends after it, and
fails without rows after a run deletion", the two `expect(...).rejects`
checks are now created immediately after the two appends start. The test
awaits them after it releases the deletion.
- No product code changed. The product code rejects correctly; the test
left one rejection without a handler for a short time.

## Verification

- `pnpm --filter @paperclipai/server exec vitest run
src/__tests__/heartbeat-run-event-atomic-append.test.ts`: 25 of 25 tests
pass, three runs in a row, no unhandled error.
- The failure before this change is the master shard named above. It
depends on which append rejects first, so I did not get a local
reproduction on demand.

## Risks

Low risk. The change is in one test. The assertions are the same.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Fable 5.2 (`claude-fable-5-2`), Claude Code, with tool use and
code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Fable 5.2 <noreply@anthropic.com>
2026-11-13 20:07:31 -06:00
DottaandClaude Opus 5.5 81d06cfc97 feat(mcp-tasks): add a demo MCP server that uses Tasks, test against it, and keep audit records out of agent errors (#21464)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents call tools in connected apps through the tool gateway, which
speaks MCP to each app
> - MCP Tasks (#21263) lets an app answer a long tool call with a
durable task. It shipped behind the experimental setting
`enableMcpTasks`, off by default
> - No public app returns tasks without credentials, so nobody had seen
the feature work against a real server, and it was hard to understand
> - This pull request adds a small demo app that uses Tasks, tests the
gateway against it over HTTP, and records a live end-to-end run
> - The live run found one bug: every task stop sent the call's audit
record, with a server stack trace, to the agent. This pull request fixes
it
> - The benefit is a repeatable way to try and test Tasks,
plain-language docs, and evidence for the decision to turn the setting
on

## Linked Issues or Issue Description

Refs #21389
Refs #21263

## What Changed

- Add an MCP Tasks demo server to `packages/kv-demo-mcp-server` (new bin
`paperclip-tasks-demo-mcp-server`; run `node
packages/kv-demo-mcp-server/dist/tasks/main.js` after a build). It is
stateless MCP 2026-07-28 with `server/discover`, and it declares
`io.modelcontextprotocol/tasks`. It has six tools: `slow_report`,
`needs_approval` (form question), `will_fail`, `cancellable`, `expires`,
and `quick_echo`. It saves task state to a file, so it can restart
mid-task. It needs no credentials. It has a status page at `/` and JSON
at `/api/state`.
- I put it in the existing demo package, so there is no new workspace
package and no `pnpm-lock.yaml` change.
- Add `server/src/__tests__/mcp-tasks-demo-server.test.ts`. It runs the
real gateway against the real demo server over HTTP, with one test per
tool plus a restart.
- Fix: `server/src/routes/tool-gateway.ts` no longer sends
`details.execution` to callers. That record has request IDs and the
server-side error stack. The audit still keeps it. Add
`server/src/__tests__/tool-gateway-agent-errors.test.ts` (red before the
fix, green after).
- Docs: `doc/connections/MCP-TASKS.md` now has "In plain words", "What a
person sees", "Try it yourself", the live-run results, and a
recommendation. The package README documents the demo server.
- `enableMcpTasks` stays off by default.

## Verification

- `packages/kv-demo-mcp-server`: `npx vitest run` gives 25/25 pass (13
new demo-server tests). `tsc -p tsconfig.test.json` is clean.
- `server`: `mcp-tasks-demo-server.test.ts` gives 8/8 pass.
`tool-gateway-agent-errors.test.ts` gives 2/2 pass, and it failed 2/2
before the fix.
- Existing suites that cover the changed route and Tasks:
`mcp-tasks.test.ts` and `tool-gateway-discovery-http.test.ts` (results
in a comment below).
- `pnpm verify` exits 2, because the doc is a global CI input. Required
CI gives the broad coverage.
- Live end-to-end run on a local `test-drive` instance with
`enableMcpTasks` on. The agent was Codex `gpt-6-luna` at low effort. The
demo was connected through "Connect your own MCP server":

| Behaviour | Result | `tools/call` count at the app |
| --- | --- | --- |
| Polling (`slow_report`, 45 s) | Delivered after 22 polls | 1 |
| Question (`needs_approval`) | The card was answered in the UI.
`tasks/update` was sent, and the task finished | 1 |
| Failure (`will_fail`) | `provider_task_failed` with the app's message
| 1 |
| Cancel (`cancellable`) | Stop on the run sent `tasks/cancel` | 1 |
| Expiry (`expires`, 15 s) | `provider_task_expired`, no hang | 1 |
| Restart (`slow_report`, 120 s) | The server was down for 2 min 23 s.
The result was delivered once, and the agent was woken once | 1 |

After the run, all 6 `tool_mcp_tasks` rows were settled and none had a
next poll.

## Risks

- Low risk. The demo server is new code in a demo package, and the bin
is additive.
- The route change removes `execution` from tool-gateway error
responses. No UI code or test reads that field from these responses. The
audit and tool-call events still record it.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), in Claude Code with tool use and
code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:47:58 -06:00
DottaandClaude Opus 5.5 1f388ed385 Start follow-up runs normally when a warm sandbox's window ends during its idle close (#21504)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - A warm sandbox keeps an agent's Runner alive between runs, so the
next run on the task starts fast
> - When a run ends, the server closes the idle warm Runner in-process.
A separate SQL worker stops the sandbox when its warm window ends
(#21410, merged today)
> - Both steps start at the same moment. The sandbox stop wins, so the
idle close fails. The failed owner stays registered, and the next run
throws `native_instruction_materialization_handoff_unavailable` before
any provider work
> - Recovery must just work: a follow-up run must never fail because a
predecessor's warm copy is gone
> - This pull request lets the idle close finish before the sandbox
stops, prepares instructions fresh when the warm copy cannot be handed
off, and releases an owner whose sandbox the host already gave up
> - The benefit is that automatic follow-up runs start normally after a
warm window ends

## Linked Issues or Issue Description

Refs #21472 (base of this stacked PR), #21410 (introduced the
warm-window stop).

**What happened?**
In a Daytona two-server chaos wave (`w-20261113T200824-s7106`), three
automatic follow-up runs failed before any provider work with
`native_instruction_materialization_handoff_unavailable`: run
`5f68e194`, run `d6bd5a97` and run `d1198e40`. None of them was at a
release swap. In each case, the predecessor's idle close had failed
first with `provider_state_unreadable` / "Daytona sandbox lease … is no
longer active". The cause: the warm-window stop from #21410 stopped the
sandbox at the same moment that the idle close started. For the first
run's predecessor, the close was requested at 20:30:34 and the sandbox
got SIGTERM at 20:30:37.8.

**What did you expect?**
The follow-up run prepares its instructions fresh and continues. The
warm-window stop does not interrupt the host's own idle close.

**Steps to reproduce**
1. Run a warm Daytona task until it ends, so its warm window starts.
2. Let the warm window end while the idle timer closes the Runner.
3. Start a follow-up run on the same task. It fails with
`native_instruction_materialization_handoff_unavailable`.

Two of the three tasks then ended stranded. That downstream part (the
claim-gate cancellation with `execution_reconciliation_required` and the
stranded notice) is owned and fixed by another session. It is **not** in
this PR.

## What Changed

- `server/src/services/native-runtime/native-session-executor.ts`
`claimWarmNativeInstructionCopy`: after it retires a stale warm owner
whose copy cannot be adopted, it returns `null`. The caller then
prepares instructions fresh. Before, it threw
`native_instruction_materialization_handoff_unavailable`.
- `server/src/services/warm-sandbox-expiry.ts`: the warm-window stop now
waits `WARM_SANDBOX_IDLE_CLOSE_MARGIN_MS` (2 minutes) past the window.
The in-process idle close (close grace plus the 30 s proof read)
finishes first.
- `native-session-executor.ts` idle timer: when an idle close fails, it
reads the lease rows for that sandbox. If every row is no longer
`active` or `retained` (the host gave the sandbox up), it releases the
in-process owner. The next run then prepares fresh.
- `doc/execution-semantics.md`: one paragraph about the stop margin and
fresh preparation.

## Verification

Red → green (each test fails on the base `35ba76d0ef` and passes on this
branch):

- `native-session-executor.test.ts` › "hands off collection before
probing or retiring a warm owner: unavailable" — red: throws
`handoff_unavailable`; green: returns `null` (fresh preparation).
- `native-session-executor.test.ts` › "forgets a warm owner whose idle
close failed only once the host gave up its sandbox: pending_cleanup /
released / retained" — red on all three; green: the owner is released
for `pending_cleanup` and `released`, and kept (retired by the
successor) for `retained`.
- `warm-sandbox-expiry.test.ts` › "leaves the host's idle close of the
warm runner time to finish before stopping its sandbox" — red: the
sandbox stops 3 s after the window; green: it stops only after the
margin. Two existing tests ("keeps a sandbox still inside its warm
window, and wakes when the window ends", "reconciles a warm sandbox left
running across a restart") now expect the deadline with the margin.

Local runs on this branch:

- `server`: `tsc --noEmit` clean.
- `vitest run src/__tests__/warm-sandbox-expiry.test.ts
src/services/heartbeat/instruction-working-copy.test.ts`: 24/24 passed.
- `vitest run
src/services/native-runtime/native-session-executor.test.ts -t "hands
off collection before probing or retiring a warm owner|forgets a warm
owner whose idle close failed"`: 10/10 passed.

Not run: a Daytona wave with this change.

## Risks

- Warm sandboxes now stop up to 2 minutes later than the configured
window. This is a small cost increase per warm sandbox. I chose a fixed
margin over an in-process "is a close running?" check because the stop
is pure SQL. The margin works when two servers run at the same time (the
server that runs the stop may not be the server that owns the Runner),
and it needs no busy-loop or cross-server signal.
- The margin does not cover a re-armed idle timer or a close that takes
longer than 2 minutes. Fresh preparation (fix 1) and the owner release
(fix 3) cover those cases.
- Not run on Daytona. The evidence is from the wave above and the unit
tests.
- In the wave, a `restore_failed` copy-back onto the just-restarted
sandbox was also seen. I did not find its root cause. It is not changed
here.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Opus 5.5 (Anthropic), model ID `claude-opus-5-5`, used through
Claude Code with tool use (shell, file edits, test runs) and extended
reasoning.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:47:42 -06:00
DottaandClaude Opus 5.5 a4c570aede feat(mcp): ask an app's form questions in the task chat and send the answer back (MCP elicitation) (#21466)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work.
> - Agents use connected MCP apps through the tool gateway.
> - An MCP app can ask the person a question in the middle of a tool
call (MCP elicitation). There are two kinds: a form, or a link to open.
> - Today a form question is a dead end:
> - A 2026-07-28 server's form `input_required` stops with "needs
information Paperclip can't collect yet".
> - An older server's `elicitation/create` makes a card, but the answer
is never sent back, and the server waits on an open stream.
> - Paperclip does not declare form support, so compliant servers never
ask.
> - This pull request shows the question as a card in the task chat. The
card names the app, and only a person can answer it. When the person
answers, the agent wakes. The agent's repeated call sends the answer to
the app exactly once, for both protocol eras.
> - The benefit: apps that need a choice or a detail from the person now
finish their work. Restarts lose nothing, and no call is left waiting.

## Linked Issues or Issue Description

Refs #21140 (stateless MCP client; this PR closes its "form elicitation"
follow-up). Refs #21263 (MCP Tasks; this PR shares its card code). Refs
#21389 (MCP Tasks follow-ups: the card answer validation item).

**Is your feature request related to a problem?**
When a connected MCP app asks the person a form question during a tool
call, the call stops and cannot continue. A 2026-07-28 server gets a
"needs input" error. An older server gets a card whose answer never
reaches it.

**Describe the solution you'd like**
Show the question in the task chat. Let a person answer, skip, or leave
it. Wake the agent when the question is resolved. Send the answer on the
agent's repeated call, once, with the server's exact state.

**Describe alternatives you've considered**
Holding the tool call open until the person answers. We rejected this:
it holds a stream and a run slot, and it does not survive deploys or
restarts. The plan (`doc/plans/2026-11-13-mcp-elicitation.md`) records
nine design decisions with their alternatives.

## What Changed

- **Plan:** `doc/plans/2026-11-13-mcp-elicitation.md` covers the spec,
the inventory, the decisions, the journey, tests, and risks.
- **New module `server/src/services/mcp-elicitation.ts`.** It is the
single mapping for direct calls and MCP Tasks:
- It turns each spec form field into a real card field: text with
defaults, number, yes/no, single choice (`enum` and `oneOf`), and
multiple choice (`items.enum` and `items.anyOf`).
  - It turns card answers back into typed `content`.
- Answered sends `accept`. Skip sends `decline`. Withdrawn, closed, or
expired sends `cancel`.
- It refuses forms that ask for passwords, keys, tokens, card numbers,
or one-time codes.
  - It keeps app text plain, so links in it are not clickable.
- It seals `requestState` with AES-256-GCM, bound to the company and
connection.
- **New table `tool_mcp_elicitations`** (migration 0479). It is
company-scoped, with one row per question round. It holds a keyed call
identity, the sealed state, the questions and their cards, a claim time,
and a 24-hour expiry.
- **Gateway, stateless servers:**
  - Paperclip declares `elicitation: {form, url}` on task tool calls.
- A form-only `input_required` saves the round, makes the cards, and
stops with a wait message.
- A repeated call before the person answers stops again without
contacting the app.
- After the answer, the repeated call claims the round and sends
`inputResponses` and the exact `requestState`, once.
- A failed send releases the claim. A crash mid-send lets the claim
lapse.
- **Gateway, older servers:**
  - Paperclip declares form support at `initialize`.
  - The first question gets `cancel` at once, so the server never waits.
- When the same question (same words and schema) arrives on the repeated
call, Paperclip answers it from the card.
- **Gateway, mixed rounds and retries:**
  - A form beside a URL step shows the link first.
- A repeated call that stopped for a question runs again in the same
run. It is no longer replayed by the per-run idempotency key.
- **Cards:** the title is "*App* needs your input". The resolver policy
is `human_only`. "Send to *App*" sends the answer. A user comment does
not supersede the card.
- **Activity log:** new events `tool_gateway.elicitation_sent`,
`tool_gateway.elicitation_refused`, and
`tool_gateway.elicitation_expired`. They record the action only, never
the values.
- **MCP Tasks:** task cards use the same mapping and gain the app name,
real fields, and `human_only`. A secret or an unsupported form stays
unsupported.
- **Storybook:** `Tasks and conversations/Questions and approvals/MCP
App Questions` has five states: Asking, Answered, Skipped, Expired, and
TwoQuestions. The fixtures are generated from the server mapper's
output.
- **Docs:** `doc/connections/MCP-ELICITATION.md` (linked from the
README) is new. `MCP-PROTOCOL-VERSIONS.md` and `MCP-TASKS.md` are
updated.

## Verification

- `server/src/__tests__/mcp-elicitation.test.ts` has 16 tests. They pass
on the branch.
- **Red/green:** 9 of the 9 gateway tests fail with master's
`tool-gateway.ts` and `mcp-http.ts`. The 7 pure unit tests cover the new
module.
- The tests use a fake 2026-07-28 server and a fake 2025-06-18 server
that asks over the open stream and waits for the reply.
- Covered: answer, skip (`decline`), withdraw (`cancel`), two questions
at once, a restart while waiting, a crash mid-send, another company with
the same app and arguments, expiry, a refused password field, an agent
that tries to answer (rejected), and capabilities.
- Existing suites: `mcp-tasks`, `mcp-protocol-versions`,
`tool-gateway-service`, `tool-gateway`, and `remote-mcp-connectors`. See
the results comment below.
- Storybook: the five stories render in headless Chromium, and the play
tests pass. The only console errors are the 404s from the backend-less
API calls that every story makes.
- `tsc --noEmit` is clean for `server` and `ui`. `node
scripts/storybook-catalog.mjs` passes.

## Risks

- **Older servers may start asking questions.** Declaring form support
at `initialize` can make older servers ask where they did not before.
Mitigation: the first question always gets `cancel` at once, so no call
hangs, and the agent gets a clear next step.
- **Per-call cost.** Each 2026-07-28 HTTP tool call in a task makes one
indexed lookup for a waiting round. Nothing is added before the first
token or to agent boot.
- **Migration.** One additive table with no backfill, so it is safe to
apply online.
- **Changed Tasks behavior.** A task card that is withdrawn or closed
now sends `cancel` instead of `decline`, as the spec defines. A skipped
card still sends `decline`.
- **Stale cards.** Expiry is checked only when the agent calls again. A
card can sit past 24 hours until the next call. Then it is withdrawn and
the app asks again.

> For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and
discuss it in `#dev` before opening the PR. Feature PRs that overlap
with planned core work may need to be redirected — check the roadmap
first. See `CONTRIBUTING.md`.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code with tool use
(shell, file edits, headless browser), with extended reasoning.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:46:56 -06:00
DottaandClaude Opus 5.5 4879a146fb fix(secrets): stop credential reads from aborting concurrent budget admissions (#21452)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Paid model calls (company decisions, fast responses, browser use,
prompt-injection checks) pass a budget admission before they start
> - Admission runs in a REPEATABLE READ transaction and share-locks the
credential's secret row, so a key rotation cannot slip between admission
and the call
> - Every credential read also wrote "Last used" (`last_resolved_at`,
`updated_at`) to that same row
> - When two calls ran at once, one call's credential read aborted the
other's admission with a serialization failure (40001). The retry then
saw the first call's new budget hold and denied the work as
`budget_blocked`
> - This is the flaky `decision-models.test.ts` failure that held Cloud
readiness, and so releases, on 2026-11-13
> - This pull request stops credential reads from writing the secret
row. "Last used" now comes from the access event that every read already
records
> - The benefit is that concurrent paid calls are no longer wrongly
denied or retried, and the release gate no longer fails on this test

## Linked Issues or Issue Description

No public issue exists. The bug is described here.

**What happened?**
Cloud readiness `verify / General tests (server (4/10))` failed on
master commits 6917d0568d and 5ede64f247. The test "admits concurrent
work from the same budget observation and records both obligations"
reported `expected "vi.fn()" to be called 2 times, but got 1 times`.
#21387 replaced the 1-second poll with an unbounded wait. That does not
fix it: one decision really is denied, so the test now waits until the
test timeout.

**Root cause**
I logged the accounting retry loop under load. Each batch of 50 runs had
6 to 10 retries with `could not serialize access due to concurrent
update` on `select … from company_secrets … for share`. The other
decision's credential read had updated the row. In 4 of 5 batches, the
retried admission saw the other decision's committed reservation and
returned `budget_blocked`.

**Expected behavior**
A credential read does not change admission. Two decisions admitted from
the same budget observation both run and both record their obligation.

**Steps to reproduce**
Run the updated test on the old `secrets.ts`. The test makes the second
credential read land after the first admission's snapshot. It then
commits an unrelated hold after both observations. The old code fails
every run with `first settled before its provider call:
{"status":"unavailable","reason":"budget_blocked"}`.

## What Changed

- `secretService` no longer writes `last_resolved_at` / `updated_at`
when it resolves a value. It still records the `secret_access_events`
row, as before.
- `secretService.list` reports `lastResolvedAt` as the later of the
stored column (legacy values) and the newest successful access event. It
reads one row per secret, backwards on the existing `(secret_id,
created_at)` index.
- A read without a consumer context (for example a GitHub API credential
read) now records a `system` access event, so it still counts as "Last
used".
- The removed `UPDATE` also made a credential read wait for a concurrent
rotation to commit. `credential()` in `ai-connections.ts` depends on
that wait to re-read the rotated version instead of handing a run a
consumed single-use token. Its existing version re-read now takes `FOR
SHARE`, so it keeps that wait. `prepareManagedAiRuntime` has its own
before/after freshness read around `runtimeCredential()`; that read
takes `FOR SHARE` too. A share lock creates no new row version and
cannot abort an admission. It adds no statement.
- The first version of this PR throttled the write to once a minute.
Review found that a read crossing the minute could still abort an
admission, so the write is removed instead.
- The concurrent-admission test is now deterministic. It starts decision
A and waits until A is blocked, then starts decision B and waits again.
It commits an unrelated 1-cent hold after both observations, then checks
that both decisions are admitted. A decision that settles before its
provider call fails the test immediately with its result, instead of
hanging.

## Verification

- Red: the new test against the old `secrets.ts` fails on every run with
the `budget_blocked` result above.
- Green: the same test with `repeats: 49` (50 runs) passes 50/50 locally
on a machine at load average 20 to 60.
- Before the change, 3 parallel batches of 50 runs of the old test each
produced a false denial.
- `npx tsc --noEmit -p server`: no errors.
- New `secrets-service.test.ts` test: a resolution leaves the secret row
unchanged, and `list()` still shows "Last used".
- `ai-connections.test.ts` quota test now checks that a read leaves
`updated_at` unchanged and records a success access event.

## Risks

- `updated_at` on a secret no longer moves when the secret is read. It
still moves on every real change.
- "Last used" on the Secrets page costs one indexed lookup per listed
secret.
- Only `list()` derives "Last used". Other endpoints that return a
secret row show the stored column, which is now set only by older
releases.
- No migration. No API change.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with tool use and
code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:46:50 -06:00
DottaandClaude Opus 5.5 e67f3bd63a test(environments): add an opt-in live Daytona proof for warm sandbox expiry (#21463)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents run in remote sandboxes, such as Daytona, through reusable
environment leases
> - #21410 made Paperclip stop a warm sandbox when its warm window ends
> - Its tests use a real database but a mocked provider stop, and
staging had no warm reusable sandboxes to observe
> - This pull request adds an opt-in live test that proves the behavior
against real Daytona
> - The benefit is a repeatable live proof of warm expiry, resume, and
recovery after a restart

## Linked Issues or Issue Description

Refs #21410

**What happened?**
#21410 has no live provider proof. Staging creates no warm reusable
sandboxes today, so the deployed change cannot be observed there.

**Expected behavior**
A repeatable live check shows that a warm sandbox stays started inside
its window and stops after it. It also shows that the next turn resumes
the same sandbox, and that a restarted cleanup worker still stops the
sandbox on time.

**Steps to reproduce**
1. Set `PAPERCLIP_LIVE_WARM_EXPIRY=1` and `DAYTONA_API_KEY`.
2. Run `pnpm exec vitest run
src/__tests__/warm-sandbox-expiry.live.test.ts` in `server`.

**Additional context**
The test makes no model turn. It drives the production environment
runtime and the cleanup worker against real Daytona sandboxes.

## What Changed

- New `server/src/__tests__/warm-sandbox-expiry.live.test.ts`. The test
is skipped unless `PAPERCLIP_LIVE_WARM_EXPIRY=1` is set. It uses:
  - the real Daytona plugin and real sandboxes;
  - `environmentRuntime.acquireRunLease` with a 2 minute warm window;
  - `releaseRunLeases(..., "keep_running")`;
- the production `registerEnvironmentCleanupWork` on a real
`createDeliveryWorkCoordinator`, which must wake itself at the deadline.
- The test checks the following:
  - The sandbox is started 20 s before the window ends.
- After the window, the worker stops the sandbox. The sandbox is
stopped, not deleted, and the lease is released as stopped and kept.
  - The second turn resumes the same sandbox (`acquisition: resumed`).
- A new database client, a freshly loaded plugin module, and a new
cleanup worker start in the middle of the window. That worker stops the
sandbox after the window ends, not before.
- At the end, the test deletes only the sandboxes whose labels match its
own company and environment.

## Verification

- Live run on 2026-11-13 against the Daytona org, window 120 s, sandbox
`170bb215…`:
  - 18:39:32.624Z turn 1 lease retained; window ends 18:41:32.624Z
  - 18:41:13Z sandbox `started` (20 s before the window ends)
- 18:41:36Z sandbox `stopped`, auto-delete -1: the worker stopped it 4 s
after the window ended
- 18:41:39Z turn 2 resumed the same sandbox (`resumed`, `started`);
window ends 18:43:39.549Z
  - 18:42:10Z restart: a new worker booted; sandbox still `started`
  - 18:43:45Z sandbox `stopped`, 5 s after the window ended
  - 18:43:45Z fixture sandbox deleted; Daytona returns 404
- Provider calls: one acquire, one resume, two stops, no destroy before
cleanup
- Without the env flag the test is skipped (1 skipped).

## Risks

- Low risk. The change adds a test file only, and the test is skipped by
default. It makes real provider calls only when someone opts in.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used through Claude
Code with tool use (shell, file edits, Daytona API).

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:42:08 -06:00
DottaandClaude Opus 5.5 e6df84a992 fix(ci): count a step cancelled as "A task was canceled." as runner loss when the log shows the shutdown (#21514)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - PR CI runs on AWS spot runners, and `ci-runner-loss-rerun.yml`
re-runs a PR's first attempt once when every failed job lost its runner
(#21291)
> - On 2026-11-13 a watcher was added to the runner image that stops the
GitHub runner about 10 seconds after AWS announces a spot reclaim, so
the lost job ends in about 28 seconds instead of up to 10 minutes
> - When the runner is stopped this way, the cancelled step's check
annotation is usually "The operation was canceled.", but a step that was
still starting up gets "A task was canceled." instead
> - The classifier only recognised the first text, so it counted such a
job as a real failure and refused the re-run, even though the job log
has the shutdown-signal line
> - This pull request recognises both cancellation texts, still only
together with shutdown evidence in the log
> - The benefit is that lost jobs stopped early in a step are re-run
automatically too

## Linked Issues or Issue Description

Refs #21291

**What happened?**
In the first hour after the spot-notice watcher went live, a `ci /
Build` job was stopped 13 s after it started. Its job log has
`##[error]The runner has received a shutdown signal...`, but its only
failure annotation was `A task was canceled.`. `classifyJob` treated
that annotation as a real failure, so the whole run was not re-run.

**Expected behavior**
A job whose log shows the shutdown signal is counted as runner loss,
whichever cancellation text the step was given.

**Actual behavior**
It is counted as a real failure, and the run stays red until a person
re-runs it.

## What Changed

- `.github/scripts/runner-loss-rerun.mjs`: the cancellation annotation
check accepts "The operation was canceled." and "A task was canceled.".
The log check still decides. A cancellation with no shutdown or
lost-communication line in the log is still a real failure, and its
evidence now quotes the actual annotation text.
- `.github/scripts/tests/runner-loss-rerun.test.mjs`: a test for "A task
was canceled." with and without shutdown evidence.
- `doc/DEVELOPING.md`: the runner-loss paragraph names both texts.

## Verification

- Red: the new test fails against the unchanged script (25 pass, 1
fail).
- Green: `node --test .github/scripts/tests/runner-loss-rerun.test.mjs`:
26 pass.
- `node --test '.github/scripts/tests/*.test.mjs'` (the `policy` job's
command): 808 pass.
- The workflow runs from `master`, so this PR's own CI cannot exercise
the new behaviour; the tests are the evidence.

## Risks

- A job cancelled as "A task was canceled." for another reason is still
treated as a real failure unless its log has the shutdown or
lost-communication line, so a person cancelling a run still blocks the
automatic re-run.
- No change to triggers, permissions, the once-only rule or the
stale-head checks.

## Model Used

Claude Opus 5.5 (`claude-opus-5-5`), via Claude Code, with tool use and
code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:41:10 -06:00
DottaandClaude Opus 5.5 01f764fe61 test(evals): pin the case clock in the original-deadline rejection tests (#21501)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The Runner support suite tests the chat catalog observers that grade
agent work
> - One observer test fails at random on unrelated pull requests
> - The test simulates a late poll result with a fixed clock jump from
the fixture time. The observer computes its deadline from a later clock
read. If 2 ms or more pass between the two reads, the jump does not pass
the deadline.
> - This pull request pins the clock to the fixture time for the
deadline rejection cases, as the sibling deadline tests do
> - The benefit is a deterministic test that still fails when the
late-deadline guard is removed

## Linked Issues or Issue Description

**What happened?**


`tests/runner-e2e/chat-catalog-domains/identity-plan-decision-observer.test.ts`
> "rejects late await original deadline without resetting it" failed in
the "ci / Runner support suite" job of PR #21418 (run 43503383519,
2026-11-13 20:38Z). The error was: `expected [Function] to throw error
including 'CONTROL_SOURCE_NOT_SETTLED' but got
'CONTROL_AFTER_SOURCE_SETTLED'`. PR #21418 changes only a plan document
and a benchmark script.

**Expected behavior**

The test passes every time. The result must not depend on how fast the
test worker runs.

**Steps to reproduce**

1. Add `await new Promise(r => setTimeout(r, 2));` before the `act` call
in the "late await" case.
2. Run `node node_modules/vitest/vitest.mjs run --config
tests/runner-e2e/vitest.config.ts
chat-catalog-domains/identity-plan-decision-observer.test.ts`.
3. The test fails with the CI error. With a gap of 0 or 1 ms it passes.

**Paperclip version or commit**

`master` at 2a8b485140.

**Relevant logs or output**

The test sets the clock to `f.now + 180_001` after the poll loads.
`f.now` is the fixture's `Date.now()`. The observer sets `deadlineAt =
Date.now() + 180_000` later, in `act`. The accept check is `Date.now() <
deadlineAt`. It is true when `act` reads the clock 2 ms or more after
the fixture. A loaded CI worker can create that gap.

## What Changed

- In the four "rejects … original deadline without resetting it" cases,
`Date.now` now returns the fixture time until the test jumps it. The
"late await" jump is then exactly 1 ms past the observer's deadline. The
other three cases fail at the budget check as before.

## Verification

- Red: with a 2 ms or 5 ms gap before `act`, the unchanged test fails
with the CI error (2 of 2 runs). With a 0 or 1 ms gap, it passes.
- Green: with the fix, the test passes with gaps of 0, 2, 5 and 50 ms
(39/39 tests each).
- Mutation: with the fix, I removed `Date.now() < deadlineAt &&` from
the accept check in `identity-plan-review.ts`. The "late await" test
fails. The assertion still detects the regression that it guards.
- `pnpm verify` selected broad CI coverage (exit 2) because the file is
a global CI input. Required CI supplies that coverage.

## Risks

- Low risk. The change affects only one test file. It does not change
product code or assertions.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), in Claude Code with tool use and
shell execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:41:04 -06:00
DottaandClaude Opus 5.5 ce5f10d534 fix(wake-queue): start a person's saved message after a legacy run is continued from saved work (#21518)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When an agent run fails and cannot resume exactly, Paperclip
continues the task in a fresh run from its last saved work
> - If a person sent a message while the task was blocked, the
fresh-continuation fallback lets that saved message start the next run
(`delivery: saved_input`) and does not queue its own run
> - The release step that promotes the saved message skips the "needs
reconciliation" hold for a native run that the fallback superseded, but
not for a legacy run
> - So for a legacy run the task moves to To do, the person's message
stays saved, and no run starts until a watchdog finds the task idle
> - This pull request applies the same superseded-run rule to legacy
runs in the release step
> - The benefit is that a person's message after a failed run starts the
next run at once, for legacy runs as for native runs

## Linked Issues or Issue Description

**What happened?**
A person sent a message on a task whose last legacy run failed.
Paperclip continued the task from saved work and moved it to To do. The
message stayed saved. No run started until the assigned-work watchdog
queued one later.

**Steps to reproduce**
1. Let a legacy run fail with an active recovery hold, so the task is
Blocked.
2. Post a comment as a person. The wake is saved as
`deferred_issue_execution`.
3. Let the fresh-continuation fallback continue the task. It records
`delivery: saved_input`.
4. Run the deferred receipt and stranded-queue passes.

**Expected behavior**
One run starts and carries the person's message.

**Actual behavior**
No run starts. The release step returns `released` because
`legacyExecutionNeedsReconciliation` is true for the superseded legacy
run.

## What Changed

- `server/src/modules/wake-queue/adapters/postgres.ts`: the release
pre-drain check does not ask for legacy reconciliation when the run is
superseded by a fresh continuation. This matches the existing native
rule in the same function.
- `server/src/services/fresh-continuation-fallback.test.ts`: a red/green
test for native and legacy runs. A person's comment goes through the
real wake path while the task is blocked, the task continues from saved
work, and exactly one run must start with the comment.

## Verification

- Red: before the fix, the legacy case fails. The saved wake stays
`deferred_issue_execution` and no run starts. The native case passes.
- Green: after the fix, both cases pass.
- `pnpm verify` for the affected selection.

## Risks

- Low risk. The change applies only when the fresh-continuation fallback
has already marked the run superseded. That decision already records
that the old run is not replayed, and the successor runs as a fresh
turn. Other callers of `legacyExecutionNeedsReconciliationWithEvidence`
(retry scheduling, recovery) are not changed, so a superseded run still
cannot be retried automatically.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Claude Code agent with tool use
and code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:40:42 -06:00
DottaandClaude Opus 5.5 f610e6a880 fix(recovery): replace a queued run the dispatcher keeps passing over instead of counting it as a pending path (#21502)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The assigned-work sweep is the backstop that makes sure every open
agent task has a live run, a pending path, or a named wait on a person
> - It counts a task with a queued run as healthy. A queued run that the
dispatcher declines on every pass is never started, so the task waits
forever and the sweep never looks again
> - On a hosted instance one agent had more than 30 queued continuations
that the claim declined on every pass. The sweep saw each task as
healthy
> - This pull request makes the sweep recognize a queued run that the
dispatcher keeps passing over, report it, and replace it once with a new
continuation when it is automatic work
> - The benefit is that an automatic task cannot stay stuck behind a run
that never starts, while a person's work is never cancelled

## Linked Issues or Issue Description

Refs #21391, #21349, #21433, #21394, #21443, #21450

**What happened?**

An agent had more than 30 queued automatic continuations
(`issue_continuation_needed`) that never started. The claim declined
each one on every dispatch pass. #21391 fixed that specific cause in the
chat-control gate. But the sweep still trusted any queued run as a
pending path. A run that the claim declines for any persistent reason
keeps the task "healthy" forever.

**Expected behavior**

A queued run that the dispatcher keeps passing over is not a healthy
pending path. The sweep reports it. For automatic work it cancels the
run through the run lifecycle and queues one new continuation in its
place. For a person's work it reports the run and leaves it.

**Steps to reproduce**

1. Queue an automatic run for a task and make the claim decline it
persistently (for example, a chat-control admission marker that does not
verify).
2. Let the run wait for more than one hour while its agent starts other
runs.
3. Run the assigned-work sweep for the task. Before this change, it
reports the task as `live_run` and does nothing.

## What Changed

- New violation `queued_run_stalled` in
`recovery/assigned-work-invariant.ts`. A queued run is stalled when all
of these are true (read from the database, on the database clock):
- it has been queued for longer than `QUEUED_RUN_AGING_MS` (60 minutes,
from `queued-run-dispatch-order.ts`);
  - it is the oldest queued run of its agent;
- its agent started more than `OLDEST_DISPATCH_SHARE` other runs after
the run became its oldest queued run: after the later of the run's aging
time and the moment the last older run of the agent started or finished.
- A stalled run is not a live run, and it does not let the task's
interruption projection ("working") count as a pending path. Its own
wake is not counted as a pending wake. A stalled run on a blocked task
still counts as live, so the block decides first.
- `listSuspects` lets a task through when its only active run is a
stalled queued run.
- `act()` handles the violation after all existing early returns:
- person work (the #21349 source-chain rule through
`listPersonOriginSourceRunIds`, which this change exports, or an
explicit user continuation): reported, never cancelled;
- a replacement that stalls again (`replacesStalledRunId` in its
context): reported, not cancelled again;
- both log one warning per run, with the run, agent and how long it
waited, and look again after `CONTINUATION_CAP_WINDOW_MS` (one hour);
- automatic work: the existing claim under the task row lock re-checks
the facts and the shared hourly cap and spacing. Inside that claim,
before the person check, it reads the run's row version (`xmin`) and the
number of wakes attached to it. Then the run is cancelled through the
run lifecycle (`setRunStatusFromLive` from `queued`, with that version
and wake count as its condition; wake cancelled, run event,
`releaseIssueExecutionAndPromote` with `suppressImmediateRecovery`). A
person's wake that joins the run after the check, or a claim that starts
it, changes one of them, so the cancel does nothing. Then one
continuation wake is queued with `previousRunId` and
`replacesStalledRunId` set to the cancelled run. If the cancel does
nothing, the unpublished claim entry is removed and nothing is queued.
- New `heartbeat/stalled-queued-run.ts` builds that cancel from the run
lifecycle functions. `heartbeat.ts` and `recovery/service.ts` pass it to
the sweep.
- Plain activity text: "Continued automatically: this task had waited
over an hour to start while its agent started other work, so Paperclip
queued it again."

### Why this signal

- #21433 keeps refused claims and claim back-off in memory only
(`declinedAgedClaimAt`, `claimBackoffs` in `heartbeat/queue.ts`).
Another server cannot read them. Run rows can be read from any server,
on the database clock.
- After #21349, an aged run ranks no lower than ordinary work, and the
oldest-run share starts the oldest ready queued run at least once in
every `OLDEST_DISPATCH_SHARE` starts. So when a run has been its agent's
oldest queued run since after the aging bound, and its agent started
more runs in that time than the share allows, the dispatcher offered the
run a slot and the claim declined it. Starts that happened while older
runs were still waiting do not count: those older runs were getting
their turns.
- A run behind a full agent sees no such starts and is left alone. Under
a slow backlog only the oldest queued run can match. So the younger
queued runs keep their place in line.
- Starts are counted only after the run reached 60 minutes of age, not
after the current server's boot. 60 minutes is longer than a release
swap overlap. So a run that only waits for a new server to reattach to
its Runners does not match. #21450 will add a reattach-phase gate before
this path.

### Why this does not set `retryOfRunId`

An earlier plan added `retryOfRunId` to the sweep's continuation so the
chat-control gate could follow it. An audit of its readers found three
changes in behavior: the admission-time chat-control check would start
to refuse sweep wakes; the claim would wait on the source run's cleanup
(a possible wait forever); and the agent prompt would call the run "a
Retry" that "re-delivers the wake". #21391 already makes these runs
claimable. So this change leaves `retryOfRunId` unset, as today. The
replacement links its source with `previousRunId`, as the sweep already
does.

### Questions from review, answered from the code

- **How often is the invariant pass triggered, and can two passes
overlap?** The `assigned-work-invariant` delivery queue
(`recovery/work.ts`, `scoped-deadline-work.ts`) runs on database-work
notifications, on each task's own deadline (`retryAt`), and on a 30 s
reconstruction of the suspect list. Passes are at least 1 s apart
(`minimumPassIntervalMs` default). In one process, passes of the queue
never overlap (`running` guard in `delivery-work-coordinator.ts`). Two
servers can sweep at the same time. Each repair is serialized by the
claim under the task row lock (`FOR UPDATE` in `claimContinuation`),
which re-reads the facts and the spacing. This change does not add a new
loop.
- **Can two runs or no run exist for one task?** Two runs: no. The
claim's spacing makes a second server back off, and the cancel happens
before the new wake (a queued run on the task would otherwise absorb the
wake). No run: there is a short gap between the cancel and the new wake,
because wake admission locks the task row in its own transaction. If the
server stops in that gap, the task has nothing pending, and the next
pass repairs it as `no_pending_path` under the same cap.

## Verification

- `server/src/__tests__/assigned-work-stalled-queued-run.test.ts` (new,
embedded Postgres, real `createHeartbeatQueue` admission and
`claimQueuedRun`):
- an automatic run that the real claim declines persistently, whose
agent started runs after it aged: reported, cancelled, exactly one
replacement, and the real claim starts the replacement;
  - a run behind a full agent: left alone, not a suspect;
  - a person's declined Retry: reported, not cancelled, one warning;
- a replacement that stalls again: not cancelled on two checks, one
warning, next look one hour later;
- two sweep instances on the same task at the same time: one
replacement;
- three aged runs queued in order, the first two get their turns slowly:
the third is not stalled when it becomes oldest, and is stalled only
after more than four further starts pass it over;
- a person's message joins the run (real wake admission) between the
claim and the cancel: the run is not cancelled, there is no replacement,
and the person's wake stays attached to the run;
- the run starts between the claim and the cancel: no cancel, no
replacement.
- First round, red before the fix: 4 failed, 1 passed (the full-agent
case passes before and after). Green: 5 passed.
- Review round: the three new tests on the previous head: 2 failed (the
backlog case and the person-join case), 1 passed (the started-run case,
which the `queued` check already covered). Green: 8 passed.
- Existing files run one at a time: `assigned-work-invariant.test.ts` 31
passed, `assigned-work-sweep-round-trips.test.ts` 2 passed,
`failed-continuation-continues-again.test.ts` 28 passed,
`recovery-wake-herd.test.ts` 35 passed,
`saved-work-recovery-overdue.test.ts` 4 passed.
- `pnpm --filter @paperclipai/server exec tsc --noEmit`: clean.
- `pnpm verify`: not completed locally (the machine is heavily loaded
and the run was stopped when the code changed). Required CI supplies the
broad coverage.

## Risks

- A task's oldest queued automatic run is cancelled and queued again at
most once per run. Its replacement starts at the back of the line. Only
a run that the dispatcher already passed over can match.
- A replacement that stalls again is only reported, so a persistent
claim refusal cannot cause repeated cancellations.
- The new subqueries run only for queued runs older than one hour. They
use the existing `(company_id, agent_id, started_at)` index. Healthy
runs and the path before `turn.start` do not change.
- #21443 (automatic-run brake) and #21450 (reattach first) touch
`assigned-work-sweep.ts`. The new path comes after every early return
that `act()` has today, so their gates can go before it.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, via Claude Code with
tool use (shell, file edits, test runs). A subagent on the same model
audited the `retryOfRunId` readers.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:40:38 -06:00
Devin FoleyandPaperclip 0a96f972e1 Connect Tailscale networks through device sign-in (#21526)
## Thinking Path

> - Paperclip manages AI agents and the environments where they run.
> - SSH environments can reach machines through a company Tailscale
connection.
> - The initial connector requires a managed OAuth app that Tailscale
limits to its own tailnet.
> - Customers need to connect their own tailnets without registering an
app or supplying a key.
> - This pull request uses Tailscale's normal device sign-in for new
connections.
> - Customers approve the Paperclip device and then select their
machines in SSH environments.

## Linked Issues or Issue Description

**Problem or motivation**

The current Tailscale enrollment path cannot serve independent customer
tailnets through the documented self-service OAuth app model. It also
requires Cloud broker configuration and pilot admission before a
customer can start.

**Proposed solution**

Start a private Tailscale device in the tenant and open its normal
sign-in link. Save its encrypted device state before reporting a
successful connection. Retain the existing SSH machine selection and
agent account behavior.

**Alternatives considered**

Auth-key entry would require customers to copy a credential. A public
OAuth integration still needs a confirmed registration path. Device
sign-in uses a documented Tailscale flow without either requirement. See
[device
authentication](https://tailscale.com/docs/features/tsnet#device-creation-and-authentication)
and [OAuth app
limitations](https://tailscale.com/docs/features/oauth-apps).

**Related PR search**

Searched GitHub PRs for `tailscale` and `device sign-in`, and issues for
`tailscale`. This follows the merged network connector in
[#20617](https://github.com/paperclipai/paperclip/pull/20617) and
preserves the SSH agent-account support in
[#21248](https://github.com/paperclipai/paperclip/pull/21248). The open
[#15648](https://github.com/paperclipai/paperclip/pull/15648) proposes
storing an OAuth client and checking its health; this change instead
enrolls a device through interactive sign-in so customers do not need an
OAuth app.

## What Changed

- New connections use device sign-in. Existing connected devices and
pending legacy OAuth callbacks remain supported.
- Pending sign-ins are bound to their company and initiating user. They
have a fixed ten-minute deadline. Device state uses server-owned vault
secrets with trusted version and digest checks.
- The active Tailscale CLI waits for consent and device approval.
Cancellation, expiry, secret deletion, and stale completion cannot
restore a cancelled connection.
- The connector stays open during sign-in. It provides a link when
popups are blocked and shows device approval separately. Progress checks
stop on completion, failure, expiry, or unmount.
- Shared response types, OpenAPI contracts, SDK source pins, and the
API/setup documentation describe the new flow. No database migration or
Cloud provider configuration is required.

## Verification

- 109 focused tests cover network processes, the real database and
encrypted vault, routes, UI, and legacy connection behavior.
- Server and UI typechecks pass.
- Follow-up `29cf296a597`: reproduced the missing OpenAPI route failure
locally. All 70 OpenAPI and Tailscale route tests and server typecheck
pass after synchronizing the API contract.
- Follow-up `79de05f38b9`: the transient-recovery regression failed
before the fix; all 20 connector UI tests and UI typecheck pass after
it. The query-error policy check reports no findings.
- The public SDK input registry check passes.
- [CI attempt
2](https://github.com/paperclipai/paperclip/actions/runs/43535594257/attempts/2)
passed on the final head, `29cf296a597e470d432b84767050bc539caa280e`: 57
successful checks, no failures or pending checks. The interrupted
canary, E2E shard, and serialized-suite jobs passed on rerun. Greptile
is 5/5 on this head with no unresolved review threads.
- `pnpm verify --explain` exits 2 because the documentation change
selects broad CI coverage. Focused local checks passed; broad CI passed
on `29cf296a597e470d432b84767050bc539caa280e`.
- Live local qualification on
`29cf296a597e470d432b84767050bc539caa280e`: native device sign-in
completed with interactive user approval. An isolated Ubuntu 24.04 ARM64
EC2 machine appeared in discovery and passed SSH pairing and workspace
probes. The assigned Claude Code agent used a Paperclip-managed
subscription, created `hello_world.txt` containing exactly `Hello,
world!` plus a newline, posted a completion comment, and marked the task
done. An independent read through the private connector verified the
bytes, remote user, and hostname. The remote machine had no stored
Claude login and no leftover temporary credential directories.
- Restarted the local Paperclip server and verified the same Tailscale
device identity, online peer discovery, successful SSH probe, and
unchanged file without new consent. Temporary public SSH ingress was
removed.
- Local qualification covers one user tailnet and the final Core
revision. Browser UI acceptance, two unrelated tailnets, a hosted
restart, and an agent task on the final staging revision are not yet
verified.

## Risks

- The enrollment method changes. It must pass the live qualification
above before broad rollout claims.
- Existing encrypted device state and the vault master key must survive
a restart or redeploy. Device expiry or removal still requires
reconnecting.
- The existing one-serving-process-per-instance constraint remains.
Concurrent replicas need durable network ownership.
- Pending setup uses bounded progress requests and an event-driven CLI
with a deadline. It adds no idle database scan.

## Model Used

OpenAI GPT-6 Astra (`gpt-6-astra`) through Codex in T3 Code, with xhigh
reasoning. Context window size was not exposed by the runtime. No
subagents were used.

## Checklist

- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] My branch name describes the change and contains no internal
ticket or instance details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation
- [x] I have considered and documented the risks
- [x] All Paperclip CI gates are green
- [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

---------

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-13 17:38:55 -08:00
DottaandClaude Opus 5.5 cefef8f11d perf(workspaces): inspect only hinted workspaces after a scoped commit (#21519)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The server cleans up execution workspaces whose task tree is done. A
background sweep (`terminal-workspaces`) finds these workspaces.
> - Every workspace write and every task status change sends a hint to
this sweep. Each hint starts a full rotation over every open candidate
workspace, and each candidate costs a recursive read of its task tree.
> - Runs write workspaces all the time, so with live runs the sweep
never stops, and its cost grows with the number of open workspaces.
> - On a hosted instance with 5 live runs, the task tree read ran 11.6
times per second. On a local copy with run churn it ran 81 to 101 times
per second.
> - This pull request inspects only the workspaces that a hint names or
reaches. Full rotations stay for startup, saved deadlines, failures and
hints without a scope.
> - The benefit is that a workspace change costs a few statements, not
one tree read per open workspace.

## Linked Issues or Issue Description

No issue exists. This describes the problem with the fields of
`.github/ISSUE_TEMPLATE/enhancement.yml`.

**What existing behavior does this improve?**

The terminal workspace sweep (`registerTerminalWorkspaceWork` in
`server/src/services/terminal-workspace-work.ts` and
`sweepTerminalWorkspaces` in
`server/src/services/execution-workspaces.ts`).

**Current behavior**

Producers send hints such as `workspace:<id>` and `issue:<id>`. The
worker ignores the scope: each settled hint wakes the queue, and the run
starts or continues a full keyset rotation over all candidate
workspaces, 50 per page with 1 second between pages. A hint that arrives
during a rotation also requests a rescan, so a new rotation starts after
the current one. Every candidate with an open task tree costs one
recursive tree read and is skipped.

**Proposed behavior**

The worker keeps the workspace ids and task ids from scoped hints. When
the queue runs only because of such hints, it calls
`sweepTerminalWorkspaces` with `targets`. The sweep then reads only
these candidates:

- the workspaces named by `workspace:` hints;
- the workspaces whose source task is the hinted task or one of its
ancestors (the tree of a workspace is its source task and its
descendants);
- the execution workspace of each hinted task.

A targeted inspection does not move the rotation cursor or its boundary
and does not clear saved deadlines. The per-workspace checks are the
same code as in a rotation.

A full rotation still runs at startup, when a saved deadline is due,
when a hint has no scope or an unknown scope, while a rotation is in
progress, and after a failed targeted inspection.

**Reason and benefit**

Background work must not grow with the backlog of open workspaces. With
4 candidate workspaces, a hint for one task tree now reads 1 tree
instead of 4. A hint during a rotation no longer starts another full
rotation.

## What Changed

- `server/src/services/terminal-workspace-work.ts`: parse scoped hints
into targets. Run a targeted inspection when only scoped hints are
pending. Keep the full rotation for startup, due deadlines, unscoped
hints, continuation pages and failures.
- `server/src/services/terminal-workspace-work.ts` also falls back to a
full rotation when more than 50 workspaces wait for inspection. It keeps
the 1-second pause between rotation pages and runs targeted inspections
between them.
- `server/src/services/issues.ts`: moving a task also hints its former
parent. Removing a task also hints its parent and its execution
workspace. Without this, a former parent's workspace could never be
reached.
- `server/src/services/execution-workspaces.ts`:
`sweepTerminalWorkspaces` accepts `targets`. A targeted page that fills
its limit requests a full rescan. and filters the candidates to the
targeted workspaces. A targeted inspection leaves the rotation state
alone. `terminalWorkspaceWorkSettled` takes `{ targeted: true }` from
the worker, and then a scoped hint does not request a full rescan.
Direct callers keep the old rescan behavior.
- `server/src/index.ts`: pass the new option through the existing
wrapper.
- Tests in `server/src/__tests__/terminal-workspace-work.test.ts` and
`server/src/__tests__/execution-workspaces-service.test.ts`.

## Verification

- New test `inspects only the workspaces a hint reaches, however many
trees stay open`: 3 workspaces with open trees and 1 done tree whose
child task is hinted. Before the change the targeted call checks 4
workspaces. After the change it checks 1, reads 1 tree, and archives the
done workspace.
- New test `inspects only hinted workspaces after a commit, and keeps
unscoped commits and deadlines on the full rotation`: before the change
the worker never passes targets. After the change a scoped commit runs a
targeted inspection, and an unscoped commit and a due deadline still run
a full rotation.
- `npx vitest run src/__tests__/terminal-workspace-work.test.ts
src/__tests__/execution-workspaces-service.test.ts
src/__tests__/isolated-task-directory-retention.test.ts
src/__tests__/workspace-delivery-cleanup.integration.test.ts
src/__tests__/git-workspace-expiration.test.ts` (in `server/`) pass.
- New tests `falls back to the full rotation when more workspaces are
hinted than one inspection reads`, `keeps the pause between rotation
pages while hinted workspaces wait` and `hints the former parent and
workspace when a task moves or is removed`. Each fails on the first
commit and passes on the head.
- `pnpm check:module-boundaries` passes.

Local benchmark. The real server ran against PostgreSQL with
`log_statement=all`. The fixture had 1 company, 60 agents, and open
tasks scaled from 150 to 5,000. Recovery re-dispatched runs to agents
that fail at once, and every run creates or updates an execution
workspace, so the hints keep coming. Each window is 180 seconds and
starts 150 seconds after the server is ready. The machine was shared.

The table gives the rate of the task tree read (`select id, status,
completed_at, cancelled_at, updated_at from issues ... WITH RECURSIVE
issue_tree`):

| Open tasks | Before (per s) | After (per s) |
| --- | --- | --- |
| 150 | 17.56 | 0.93 |
| 1,500 | 109.83 | 5.16 |
| 5,000 | 113.82 | 4.40 |

Total statements and transactions per second depend mostly on how fast
the failing runs cycle in the fixture, so they are noisy. Before →
after: 343 → 287 at 150, 1,647 → 1,562 at 1,500, 1,530 → 1,678 at 5,000
statements/s. In that window the 5,000-task run after the change also
had 10% more run events. With no activity, the server issued 0
statements before and after the change.

## Risks

- A workspace that waits on something without a hint is no longer
inspected as a side effect of unrelated activity. The known cases keep
their path: a workspace held by an active run keeps its past-due
deadline, so the 5-minute deadline rotation inspects it again; cooldown,
compaction, Git idle and retry deadlines still start full rotations. A
delivery that changes outside Paperclip without a work product change
was already only found by a later rotation.
- A failed targeted inspection requests a full rotation, so a lost
target is recovered by the rotation.
- Low migration risk. No schema change.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:04:35 -06:00
DottaandClaude Opus 5.5 3cd671575d perf(attention): scope native finalization retention hints to the run's task graph (#21521)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The server archives old attention items (reviews, blocked tasks,
decisions, failed runs) in a background queue, `attention-retention`.
> - A hint for a whole company makes this queue reload every attention
candidate of the company, 100 per page. Each page also reads review
attention, blocker graphs and source evidence.
> - Each native run finalization sent such a company hint, although a
run changes only the attention items of its own task graph.
> - On a hosted instance with about 1,400 open tasks and live native
runs, the decision retention read ran 26.7 times per second. Native run
finalization is one of the producers that sent a company hint on every
run.
> - The same finalization already sends the issue-scoped hint through
`notifyNativeStatusReassessment` when the run has a task or a run id.
This pull request sends the company hint only when the run has neither.
> - The benefit is that a finished run reloads only the candidates of
its task, its ancestors and the tasks it blocks, not every candidate of
the company.

## Linked Issues or Issue Description

No issue exists. This describes the problem with the fields of
`.github/ISSUE_TEMPLATE/enhancement.yml`.

**What existing behavior does this improve?**

The attention retention hint that native run finalization sends
(`notifyNativeFinalizationWork` in
`server/src/services/native-finalization-notifications.ts`).

**Current behavior**

`notifyNativeFinalizationWork` calls `notifyAttentionRetentionWork(tx, {
companyId })`. The `attention-retention` queue has no reconstruction
interval, so each finalization reloads all eight candidate kinds of the
company, page by page, with the batch eligibility reads for each page.

**Proposed behavior**

`notifyNativeFinalizationWork` sends the company hint only when the
scope has no `issueId` and no `runId`. Otherwise
`notifyNativeStatusReassessment`, which the same function already calls,
sends `notifyIssueAttentionRetentionWork(tx, { companyId, issueId, runId
})`. That helper resolves the task of the run, adds its parents and the
tasks it blocks (at most 100), and sends a hint with these task ids.
When the run has no task, or the graph has more than 100 tasks, the
helper sends the company hint itself.

**Reason and benefit**

Background work must not grow with the backlog of open tasks. The
attention items that a run can change are the failed run, the review,
the recovery actions, the decisions and the blocker items of its task
graph. The issue-scoped hint covers all of them, and the ordinary task
write path uses the same helper.

## What Changed

- `server/src/services/native-finalization-notifications.ts`: send the
company attention retention hint only for a finalization without a task
or run.
- `server/src/__tests__/attention-retention-work.test.ts`: new test
`scopes native run finalization to the task graph of the run`.

## Verification

- The new test sends the finalization hint once with a run id and once
with a task id. Before the change both hints are `{ companyId }`. After
the change both hints name the task and its parent.
- `npx vitest run src/__tests__/attention-retention-work.test.ts` and
the native work tests (`native-result-commit-work`,
`native-environment-settlement-work`, `native-run-projection-work`,
`native-recovery-notice-work`, `retained-native-session-cleanup-work`,
`native-status-reassessment-work`, `muse-terminal-settlement`,
`ownership-reconciliation-report`) pass (in `server/`).
- `pnpm check:module-boundaries` passes.

## Risks

- No new reads. The issue-scoped hint was already computed in the same
transaction by `notifyNativeStatusReassessment`.
- An attention item outside the task graph of the run is not reloaded by
this hint. Agent status, budget and company changes keep their own
hints.
- Low risk. No schema change.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [ ] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:04:30 -06:00
DottaandClaude Opus 5.5 a82fa96540 perf(run-events): collect deltas that arrive while a batch commits into one next batch (#21520)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The control plane port commits streamed text deltas of a run in
batches (#21460): the first text at once, later deltas after at most 250
ms
> - Each batch is cut when its timer fires, even when the previous batch
of the same run has not committed yet
> - When the database is slow, every run then queues one small batch per
250 ms behind its slow commit. The number of statements stays the same
exactly when the database has the least room for them, and the queue
grows
> - This pull request cuts a batch only when the previous batch of the
run has committed. Deltas that arrive while a batch commits all join the
next batch
> - The benefit is fewer, larger statements under load, with no change
when the database keeps up, and no longer wait for any delta

## Linked Issues or Issue Description

Refs #21460 (batched deltas). Refs #21511 (one statement per delta
batch). Refs #21418 (agent scaling plan and load model).

**What existing behavior does this improve?**
How `PaperclipControlPlanePort` groups queued text deltas into batches
when a batch commit is slower than the batch time.

**Subsystem affected**
Server run event ingest:
`server/src/services/native-runtime/paperclip-control-plane-port.ts`.

**Current behavior**
`#flushStreamingEvents` takes the queued deltas when it runs (on the 250
ms timer, at 100 deltas, or before any other event) and chains their
commit behind the previous batch. If one commit takes two seconds, the
run queues about eight batches, each one statement, behind it.

**Proposed behavior**
`#flushStreamingEvents` chains one cut behind the previous batch and
takes the queued deltas when that cut runs. While a cut waits, later
flush requests join it. A run has at most one batch committing and one
cut waiting.

**Reason and benefit**
Under load, the number of delta statements per run falls to about one
per commit time instead of one per 250 ms. When commits are faster than
250 ms, nothing changes.

**Breaking changes**
None. Rows, receipts, order and the first-text rule are the same.

## What Changed

- `#flushStreamingEvents` queues one cut on the run's commit chain
(`#streamingFlushQueued`). The cut takes every delta queued at that
time, sets the first-text flag, commits the batch, and then releases the
pending count of that batch.
- A flush while a cut waits returns the same chain, so a caller that
waits for it (every other event, `checkpointSession`, `completeRun`,
`replayEvents`, `provesNoRunProviderWork`) still waits for every delta
queued before it.
- A batch can now hold more than 100 deltas when commits are slow. The
100-delta rule still asks for a flush; the batch is one statement with a
fixed number of bind parameters (#21460, #21511).

## Verification

- Red/green: new port test "collects the deltas that arrive while a
batch commits into one next batch, before the next other event"
(`paperclip-control-plane-port.test.ts`). A transaction holds the run
row, so the first batch waits. Six more deltas arrive 15 ms apart with a
5 ms batch time, then a non-streaming event. Before: 8 inserts (one per
timer). After: 3 inserts (the waiting batch, one batch of six, the other
event). Rows are stored 1–9 in order and every receipt resolves.
- The whole port suite with its conformance suite passes: 101 tests.
Other suites, one file at a time: `native-observer-replay` (1),
`paperclip-semantic-conformance` (1), `native-interaction-bridge` (7),
`chat-native-progress-work` (35). I ran the four streaming port tests
two more times; they pass each time.
- `pnpm --filter @paperclipai/server run typecheck` passes. `pnpm
check:module-boundaries` passes.
- Load model (`scripts/benchmark-agent-scaling-load.ts` as merged in
#21418). The model applies the port's batching rules itself, so for
these runs I changed my local copy only (not in this pull request): with
`DELTA_COALESCE=1`, `flushDeltas` queues one cut on the agent's append
chain and takes the queued deltas when the cut runs, as the port now
does. All rows run with #21511. Default pool of 10 connections; embedded
PostgreSQL on the same, shared machine; runs one after the other with a
pause between them. No run had a failed operation, a failed checkpoint
save or a lost lease.

| Agents | Batching | Window | Load avg | Event-loop utilization |
Statements/s | Delta lateness p50 / p99 (ms) | Other events done /
offered | Other event lateness p50 / p99 (ms) |
  | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 200 | this change | 60 s | 9.7 | 0.51 | 3,832 | 151 / 3,087 | 12,000 /
12,000 | 3 / 3,994 |
| 400 | cut on timer (before) | 60 s | 12.7 | 0.74 | 5,465 | 1,129 /
6,040 | 18,719 / 24,000 | 5,721 / 15,238 |
| 400 | this change | 60 s | 12.5 | 0.78 | 7,010 | 892 / 3,131 | 23,993
/ 24,000 | 1,896 / 4,193 |
| 600 | cut on timer (before) | 60 s | 10.0 | 0.69 | 6,158 | 2,357 /
9,483 | 18,771 / 36,000 | 2,280 / 30,985 |
| 600 | this change | 120 s | 11.4 | 0.72 | 6,730 | 1,147 / 6,335 |
61,158 / 72,000 | 11,678 / 21,212 |
| 600 | this change (repeat) | 120 s | 13.6 | 0.74 | 6,498 | 1,225 /
6,057 | 58,984 / 72,000 | 11,465 / 25,361 |
| 800 | this change | 60 s | 11.5 | 0.80 | 6,228 | 2,054 / 7,186 |
24,877 / 48,000 | 24,544 / 31,331 |

The process does more work per second with this change (more other
events finish), so statements per second rise; delta lateness falls. 600
agents are still not carried reliably: other events fall behind. #21528
removes two statements from each such event and one from each checkpoint
save; its description has the stacked results.

## Risks

- A delta never waits longer than before: it already waited for the
previous batch of its run. When the database keeps up, batches are the
same as before.
- Under load, one batch can be large. The live view then gets one larger
step instead of several small steps behind the same wait.
- A failed batch now rejects more deltas at once (every delta that
joined it). The runner resends them, as before; each resend is
idempotent.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 19:00:08 -06:00
DottaandClaude Opus 5.5 0bba04675c perf(run-events): append in-order native events in one statement (#21511)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - One server process stores every event of every running native agent
in `heartbeat_run_events`
> - After #21460, text deltas commit in batches, but each batch still
takes five statements: BEGIN, the run lock, a replay probe, one insert,
and COMMIT. A single event without a wake takes four
> - In the agent scaling load model, the process issues about 36
statements per agent per second and saturates its one thread between 200
and 400 agents
> - This pull request writes a batch of in-order native events, or one
such event, with one statement that commits on its own
> - The benefit is about half the statements per agent, less CPU per
agent on the main thread, and less time on a pool connection, with the
same rows, receipts, cursor and locks

## Linked Issues or Issue Description

Refs #21460 (batched deltas). Refs #21435 (source cursor). Refs #21418
(agent scaling plan and load model).

**What existing behavior does this improve?**
The append of native source events that continue their source's stored
cursor: batched text deltas, the first delta after another event, and
every other native event that registers no wake.

**Subsystem affected**
Server run event ingest: `server/src/services/heartbeat-run-events.ts`,
and the port comment in `paperclip-control-plane-port.ts`.

**Current behavior**
A delta batch takes BEGIN, a run lock read (`for no key update`), a
replay probe, one insert, and COMMIT: five statements. One delta that
commits at once takes four. In the load model at 200 agents the process
sends about 6,700 statements per second at event-loop utilization 0.72.

**Proposed behavior**
The same append is one statement. Its update of the run row takes the
same row lock. It requires the stored cursor to end just before the
batch, requires the `next_event_seq` that its own snapshot read, and
finds no replayed event. Then it moves the sequence and the cursor and
inserts every row. If any check fails, it writes nothing and the append
takes the existing locked transaction.

**Reason and benefit**
Fewer statements free the main thread and the shared pool for other
agents. In the load model at 200 agents, statements per second fall from
about 6,700 to about 3,300 and event-loop utilization falls from 0.72 to
0.56.

**Breaking changes**
None. No schema change. Rows, receipts and the cursor are the same.
Servers from before this change keep working on the same runs.

## What Changed

- `appendNewNativeSourceEvents` (`heartbeat-run-events.ts`) is one
statement: a CTE reads `next_event_seq` from the statement snapshot, an
`update heartbeat_runs ... returning` locks the run row and moves
`next_event_seq` and the source cursor, and an `insert ... select from
jsonb_array_elements($9)` writes every row. The text is fixed, with nine
bind parameters for any batch size.
- **Why the `next_event_seq` check.** A statement that waits for the run
row lock keeps the snapshot it took before the wait. Its replay probe
cannot see events that the lock holder committed. Every event append
(native, ordinary, the run coordinator store, the retry queue) moves
`next_event_seq` in its own transaction. So the update requires the
value that its snapshot read. PostgreSQL re-checks that condition on the
newest row version after the wait, and any append committed meanwhile
makes the statement write nothing. The unique indexes on `(run_id,
source_event_id)` and `(run_id, source_instance_id, source_seq)` are a
second guard: a unique violation from this statement also falls back.
- A source without a cursor starts at 0. The probe still finds any
stored event of that source, so a run that an older server wrote falls
back and recounts.
- `appendHeartbeatRunEventBatch` and `appendHeartbeatRunEvent` try the
one statement first for native events without a wake, outside a caller's
transaction. Every other case (a replay, a gap, an event with a wake, a
cursor that lags, a run changed meanwhile, a binding mismatch, a run
deleted meanwhile, a caller's transaction) takes the existing
transaction, unchanged.
- Inside a caller's transaction the append keeps its SAVEPOINT path. A
failed statement there would abort the caller's transaction, and the
fallback could not run.
- `assertWritable` runs before the statement is sent and after it
returns (see Risks).

## Verification

- Red/green on embedded PostgreSQL. I ran the new tests against the
previous `heartbeat-run-events.ts` and saw 9 of them fail; all 25 tests
in the file pass with the change
(`server/src/__tests__/heartbeat-run-event-atomic-append.test.ts`):
- One statement for an in-order event and for a batch (before: 4 and 5).
- A writer revoked before the send sends nothing; a statement that fails
on its third row stores no row and no sequence; the resend commits.
- A writer revoked while the statement runs: the rows stay, the receipt
is rejected, and the resend is a replay with the same rows.
- A batch waits for Stop on the run row and appends after it; a batch
and a single event that wait for a run deletion fail with
`heartbeat_run_event_binding_mismatch` and store nothing.
- An older server stores events 2 and 3 while it holds the run row,
without the cursor: the waiting batch 2–4 falls back, returns duplicate,
duplicate, committed, and the cursor ends at 4.
- Two connections append the same 200-event stream in batches of 10:
each event is stored once, in source order, sequences 1–200 without a
gap, one `committed` and one `duplicate` receipt per event.
- A 7,000-event batch (more than any bind parameter limit at 10
parameters per row) is one statement. Before: it fell back to one
statement per event and timed out at 15 s.
- A run without a cursor starts at its first event; a run with events
from an older server replays and then continues.
- Equivalence: 12 random streams of 30 steps (in-order batches, gaps,
replays of earlier ranges, fills of gaps, ordinary events, conflicting
replays). Each step goes to two runs, one through the new path and one
forced through the one-by-one locked path. Receipts, errors, rows,
`next_event_seq` and cursors are equal at every step.
- Mutation check: with the `next_event_seq` check and the
unique-violation fallback removed, the older-server race test fails.
- The port suite (`paperclip-control-plane-port.test.ts`, with the
package conformance suite) passes: 125 tests. Two tests changed: the
streamed-event test now expects one statement and the new revocation
behavior, and the batch test checks that the five deltas are one
statement sent before the boundary event.
- Other focused suites pass, one file at a time:
`heartbeat-run-event-sequencing` (5), `native-event-boundary-golden`
(1), `native-observer-replay` (1), `paperclip-semantic-conformance` (1),
`native-interaction-bridge` (7), `chat-native-progress-work` (19),
`heartbeat-native-restart-suspension` (115), `live-recovery-work` (51).
In 2 of 3 runs with this change, all 51 tests passed but the file's
`afterAll` (closing the side pools and the embedded database) hit its 30
s timeout on this busy machine; the third run passed cleanly, and one
run on `origin/master` passed. The native events in that file all
register a wake, so they do not take the new path.
- `pnpm --filter @paperclipai/server run typecheck` and `pnpm
check:module-boundaries` pass.
- Load model: `scripts/benchmark-agent-scaling-load.ts` as merged in
#21418 (per-source counts stop when the window closes). I did not change
its batching rules for this pull request. Model rates: 20 deltas/s, 1
other event/s, 0.3 API requests/s per agent; default pool of 10
connections; embedded PostgreSQL on the same, shared machine. I ran
"before" (`origin/master`) and "after" one after the other, with a pause
between runs, and I give the load average at the start of each run. The
machine was busy with other work (load 10–15 on 18 cores), so compare
rows that ran close together. No run had a failed operation, a failed
checkpoint save or a lost lease.

| Agents | Code | Window | Load avg | Event-loop utilization |
Statements/s | Delta lateness p50 / p99 (ms) | Other events done /
offered | Other event lateness p50 / p99 (ms) |
  | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 200 | before | 60 s | 9.6 | 0.56 | 7,043 | 283 / 3,368 | 12,000 /
12,000 | 278 / 3,895 |
| 400 | before | 60 s | 11.8 | 0.59 | 7,158 | 9,927 / 18,779 | 6,068 /
24,000 | 1,904 / 60,154 |
| 400 | after | 60 s | 12.7 | 0.74 | 5,465 | 1,129 / 6,040 | 18,719 /
24,000 | 5,721 / 15,238 |
| 600 | before | 60 s | 15.0 | 0.85 | 12,419 | 7,108 / 11,017 | 7,836 /
36,000 | 10,822 / 52,171 |
| 600 | after | 60 s | 10.0 | 0.69 | 6,158 | 2,357 / 9,483 | 18,771 /
36,000 | 2,280 / 30,985 |
| 600 | after, `DELTA_BATCH_MS=0` | 60 s | 10.8 | 0.68 | 8,014 | 36,135
/ 52,222 | 36,000 / 36,000 | 698 / 3,422 |

At 200 agents an earlier pair gave 0.61 → 0.43 event-loop utilization
and 7,051 → 3,210 statements/s. One process with this change alone still
does not carry 600 agents: at 400 and above the 10-connection pool and
the shared database are the limit (event-loop utilization stays below
0.75 while other events fall behind). #21520 and a third pull request
for the remaining statements of other events and checkpoint saves
continue from here; their description has the stacked results.
- Statements: a delta batch 5 → 1; one delta or one native event without
a wake 4 → 1. Unchanged: an event with a wake (6) and a checkpoint save
(5).

## Risks

- **`assertWritable` window (read first).** Before, the writer check ran
inside the transaction, before the first statement and before COMMIT. A
revocation while the statements ran rolled the append back. Now a
revocation while the one statement runs does not roll it back: the rows
commit, and the caller gets the error instead of the receipt, so no
observer runs and the runner gets no ack. This check guards a controller
that hands its run to a successor at restart
(`nativeRunsDetachingForRestart`). I think this is acceptable: (1) the
earlier check was not atomic with COMMIT either, so a revocation during
the COMMIT round trip already committed; the window grows from one
COMMIT round trip to one statement round trip plus its wait for a pool
connection; (2) the append is idempotent by source event id and payload
hash, so the successor sees the committed row as a replay of the event
that the runner resends; (3) the old controller never acts on the
rejected receipt. A new test covers exactly this.
- A crash loses nothing new: the statement commits or not as a whole,
and a resend after an unknown outcome is a replay.
- The check on `next_event_seq` makes the statement write nothing
whenever any other event of the run committed after its snapshot. Under
heavy contention on one run, more appends take the fallback. The
fallback is the previous code path.
- The statement text is fixed SQL sent through the driver
(`$client.unsafe`), like the rest of the event stream statements.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 18:59:01 -06:00
DottaandClaude Opus 5.5 2178c3df88 fix(native): keep a slow but live Runner on retry instead of replacing it (#21462)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When a recovery read to a remote Runner goes unanswered, the next
attempt of the same run carries `replaceUnansweredRunner`
> - That attempt verified the Runner was alive and then refused to adopt
it anyway. On a resumed sandbox lease the reuse check then failed with
`resumed_not_suspended` and left the Runner running. Otherwise a second
Runner was launched, and its launch fence killed the busy Runner, which
lost the provider turn.
> - After a late reconnect a Runner is often slow, not dead: it is still
replaying the events it saved with no controller attached
> - This pull request makes the attempt check progress first. It adopts
a live Runner that is making progress and replaces only a Runner that
made no progress.
> - The benefit is that a slow but working Runner keeps its turn, and
the retry no longer causes the failure it was meant to fix.

## Linked Issues or Issue Description

**What happened?**

Hosted staging, release swaps on 2026-11-13. The retry after a
first-attempt timeout failed with these codes (counts from the traced
sample of 136 failures):

- `runner_ingress_unavailable: runnerd process failed:
runner_harness_state_mismatch: resumed_not_suspended` (part of 40
`ingress` failures)
- `runner_remote_process_identity_unavailable` (25): the Runner was gone
by the time the retry checked, because the launch fence or a previous
retry had killed it
- "restored the conversation after process loss, but…" /
`provider_turn_lost_on_restore` (10): the replacement Runner restored
from the old state and the provider turn was gone

Runs that crossed a swap with a reconnect gap under 150 s: 0 of 9
failed. Over 420 s: 14 of 15 failed. The process ID of the Runner
changed inside the same sandbox in the failed runs, which shows the
replacement.

Code path: `nativeRecoveryCommandUnanswered` grants an extra attempt
with `replaceUnansweredRunner: true`. `verifyRemoteRunnerOrSettledExit`
verifies the Runner is alive and then returns `undefined` because of
that flag, so `adoptedProcess` is `undefined`.

**Expected behavior**

A live Runner that is making progress is adopted. A Runner that made no
progress is stopped with proof and then restored from its own journal.

**Steps to reproduce**

1. Reattach a remote run whose previous attempt left a recovery read
unanswered (`replaceUnansweredRunner: true`), with the Runner alive and
its saved event sequence past the controller's last acknowledgement.
2. On master the attempt launches a second Runner, or fails with
`resumed_not_suspended` on a resumed sandbox lease.
3. The new tests in `native-session-executor.test.ts` and
`native-runner-restart-recovery.integration.test.ts` do this.

**Paperclip version or commit**

`master` at 426e8a35dc.

## What Changed

- `native-session-executor.ts` (`verifyRemoteRunnerOrSettledExit`): with
`replaceUnansweredRunner`, the attempt adopts the live Runner, launching
nothing and signalling nothing, when:
- the Runner's saved event sequence is past the controller's last
acknowledgement, or
- its saved state (event and command sequences, generation, lifecycle)
changes within a bounded wait of about 10 s.
- The baseline is the first successful state read; every successful read
is also compared against the controller's acknowledgement. A Runner
whose state cannot be read at all during the wait is adopted, not
replaced: missing evidence is not proof of a wedge. A Runner whose
identity changes during the wait fails closed.
- A settled exit during the wait keeps the attempt allowance that
admitted the replacement, so it also restores on the extra fourth
attempt.
- Only a Runner that made no progress is replaced. The relaunch's
existing launch fence stops it and its process lineage with proof, and
the new Runner restores its own journal. A resumed sandbox lease no
longer refuses that relaunch with `resumed_not_suspended`.
- A Runner that exits during the wait follows the existing exited-Runner
rules: a settled suspension restores, and a mid-turn exit stays
fail-closed.
- The wait runs only on this recovery path, never on a normal run start.
- `doc/execution-semantics.md`: describes adopt versus replace.
- The claim flag `replaceUnansweredRunner` is set in claim logic that
another session owns. This PR does not change it; it changes only what
the executor does with it.

## Verification

Red on master, green on this branch:

- `native-session-executor.test.ts`
- `adopts a live Runner that emitted events past the controller's last
acknowledgement`
- `adopts a live Runner whose saved state advances within the bounded
wait`
- `hands a live Runner that made no progress to the relaunch, whose
launch fence stops it, without signalling it here`
- `restores from the journal of a Runner that suspended and exited
during the wait`
- `stays fail-closed for a Runner that exited mid-turn during the wait`
- `native-runner-restart-recovery.integration.test.ts`: `reattaches a
live %s Runner whose recovery read went unanswered on a resumed sandbox
lease` (progressing, wedged)
- Review fixes, `native-runner-restart-recovery.integration.test.ts`
(the first four were red before the fix):
- `adopts a live Runner whose first state read failed and whose later
reads advance`
- `adopts a live Runner whose first readable state is already past the
controller's acknowledgement`
- `does not replace a live Runner whose state could not be read at all
during the wait`
- `restores from the journal of a Runner that suspended and exited
during the wait on the admitted fourth attempt`
- `fails closed when the Runner's identity changes during the wait`
(pins existing behavior)
- `native-remote-retry-admission.test.ts`: unchanged, still passes.

Local runs on this branch: `tsc --noEmit` is clean for `server` on this
branch (master c6b497d831, which contains #21442).
`native-runner-restart-recovery.integration.test.ts`, the
adopt-or-replace block: 12 of 12 pass. The full
`native-session-executor.test.ts` and
`native-remote-retry-admission.test.ts` runs did not finish on this
shared machine (load 20 to 96); CI ran them on this head and they
passed.

Conflict resolved: master added a relaunch branch for a Runner that
exited for a fresh bootstrap (#21374) at the same place. Both branches
are kept; the cheaper check for an unanswered Runner that made no
progress comes first.

Daytona results that exist. They ran on a combined branch with this PR,
#21450, #21442 (merged) and #21362 (merged), so they cannot be credited
to this PR alone:

- G5 baseline (master d1d71a7755; 480 s delay injected on the first
identity write per run): one Claude run reproduced the staging chain:
authentication 419 ms, identity +423.5 s, attempt 1 `native session
provider recovery timed out after 900000ms`, then the retry failed in 6
s with `runner_harness_state_mismatch: resumed_not_suspended`.
- G5c (same cell, combined branch 588c6ba267): 6 of 6 tasks finished
hands-off. The chain did not recur. All runs that crossed swap 2 were
adopted on the same Runner process ID.
- Caveat: the adopt-or-replace retry path that this PR changes was never
exercised in a cell. In G5c no first attempt failed, because all
identity writes started together 4 to 5 s after the claim. G5c also ran
before the phase timing fix in #21450.

## Risks

- Not proven on Daytona: the adopt-or-replace path itself (see the
caveat above).
- Without #21442 (merged) an adopted Runner can still time out on its
first command while it replays a backlog. This PR stops the replacement;
#21442 (merged) makes the adopted Runner answer.
- A Runner whose state cannot be read during the whole wait is adopted.
If it is in fact wedged, the adopted attempt times out again and the
attempt limit bounds it.
- A wedged Runner that still emits a few events counts as progressing
and is adopted. The existing attempt limit bounds this.
- The progress check adds a wait of up to about 10 s on this recovery
path only. A normal run start has no added wait.
- The claim construction (`native-restart-recovery.ts`) is owned by
another session and is unchanged.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 18:56:11 -06:00
DottaandClaude Opus 5.5 47a2c82d35 fix(heartbeat): reach the previous server's live runs first after a release swap (#21450)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - During a release swap, the outgoing server stops and the incoming
server takes over the live runs on remote Runners
> - The incoming server reached those runs late: it waited up to 60 s
for the outgoing server's controller lease to lapse, re-verified each
run's whole log from object storage, and deferred a reattach when the
agent directory was busy. Queued starts competed with the reattaches for
the same resources.
> - On hosted staging, the later a live run was reached, the more likely
it was to fail. Runs reached in under 150 s never failed; runs reached
after 420 s almost always failed.
> - This pull request hands live runs over at the start of shutdown,
makes a reattach skip work that the same run already did, and holds
queued starts while handed-over runs are being reached, with a 180 s
cap.
> - The benefit is that live runs are reached within seconds of a swap
instead of minutes, so they continue instead of failing.

## Linked Issues or Issue Description

**What happened?**

Hosted staging, release swaps on 2026-11-13 (06:05Z to 09:45Z). For each
run that crossed a swap we measured the gap from the last Runner output
to the moment the new server recorded the Runner's process identity:

- gap under 150 s: 0 of 9 runs failed
- gap over 420 s: 14 of 15 runs failed
- median gap: 180 s for runs that resumed, 470 s for runs that failed

In one traced run the incoming server waited on
`remote_controller_lease_live` three times (09:14:21, 09:15:18,
09:16:17), claimed at 09:16:59, and spent 85 s in `task.prepare`. Of
that, `heartbeat.prepare.run_output` took 58.7 s, because it re-read the
run's whole log from object storage. A control run spent 23 s in
`task.prepare`. Queued starts used the same preparation path, the same 8
object-I/O slots and the same database pool, and nothing gave a reattach
priority. The first-attempt failures were `PRP command session.snapshot
timed out` and `codex_history_read_failed`. The retry then failed with
`runner_harness_state_mismatch: resumed_not_suspended`,
`runner_remote_process_identity_unavailable` or
`provider_turn_lost_on_restore`.

**Expected behavior**

Every live run is reached within about 30 s of the swap. Queued work
starts after that, or after a bounded wait.

**Steps to reproduce**

1. Run a two-server release swap with live remote runs and queued runs.
2. Observe `native.recovery.transition` "waiting" events with
`remote_controller_lease_live`, then the claim, then a long
`heartbeat.prepare.run_output` span.
3. The new tests in `reattach-phase.test.ts` and
`heartbeat-native-early-hand-off.test.ts` model this.

**Paperclip version or commit**

`master` at 426e8a35dc.

## What Changed

- Early hand-off at shutdown (`heartbeat/recovery.ts`,
`server/src/index.ts`): in deploy-overlap mode, the outgoing server
releases the remote controller lease of its live runs at SIGTERM
(`recoveryState` `awaiting_runner_reattach`). A run another server
already took over is left alone. The incoming server no longer waits for
the lease to lapse.
- Reattach fast path:
- `heartbeat/run-output.ts`, `native-runtime/native-run-log-export.ts`:
a reattach starts without waiting for the existing log to be
re-verified.
- `heartbeat/agent-directory-busy.ts`, `heartbeat.ts`: a reattach that
meets a busy agent directory waits in process (at most 30 s) instead of
being deferred to a full re-preparation. An ordinary start is still
deferred at once, with no added wait.
- Reattach phase lifecycle (`heartbeat/reattach-phase.ts`): the watch is
a worker on the shared delivery coordinator, queue `reattach-phase`.
- Trigger: one startup listing of a previous server's live runs; then
the hand-over notification (`notifyReattachHandOver`, sent in the
hand-over transaction); the earliest lease deadline it watches (a killed
owner); and a 2 s re-check only while handed-over runs are still
pending, which ends with the phase (cap 180 s). The bounded re-check
exists because a run can also end on another path that sends no
notification to this queue.
- Source of truth: `native_run_finalizations` and `heartbeat_runs` rows.
- Quiet condition: nothing pending and no watched lease, so no deadline
and no SQL. It stops with the coordinator at shutdown.
- Reached hook for every driver (`native-session-executor.ts`): a
reattached run counts as reached when the native session runtime hands
the executor its recovered or opened session (`onSession`). Every driver
reaches that point only after the Runner answered this server's attach
and recovery commands, including an adopted `RunnerdExternalDriver`
Runner, which records no new process identity. The identity hook stays
as an earlier signal.
- Claim wake: when the reattach-phase worker finds runs that just became
claimable (hand-over notification or a killed owner's lease deadline),
it starts this server's restart-recovery claim for them at once. A
validation cell showed the claim otherwise waited about 21.5 s after the
hand-over.
- `recovery/assigned-work-sweep.ts`: the `deferred_reattach_phase`
return now sits before every action path of `act()`, except starting an
overdue saved-work recovery (it resumes an existing run).
- The in-transaction admission re-check of a queued claim
(`confirmRunAdmissionStillOpen`) also honours the reattach phase. A
phase that begins during a claim returns the run to the queue through
the existing rollback.
- A reattach whose log re-verification runs in the background awaits it
before the log is finalized (`settleDeferredLogRecovery`, both
finalization sites in `run-completion.ts`).
- Reattach phase (`heartbeat/reattach-phase.ts`,
`heartbeat/run-admission.ts`): `isReattachPhaseActive()` is true while
runs that a previous controller handed over (lease released or observed
expired, or claimed by this server's restart recovery) are claimable but
not yet reached. A run is reached when its process identity is recorded.
Authentication is logged as a mark. The cap (180 s,
`PAPERCLIP_REATTACH_PHASE_CAP_MS`) counts from the moment the runs
became claimable. Later batches join or re-arm the phase. Runs reached
after 150 s are counted and logged.
- Gates on the phase:
- `claimQueuedRun`, start-next, resume-queued and dispatch in
`heartbeat/queue.ts`, `run-lifecycle.ts` and
`native-runner-execution-driver.ts` leave queued runs queued.
- `execution-recovery-resolution.ts`: one line, the first line of
`deliverReconciledExecutions`.
- `recovery/assigned-work-sweep.ts`: one line, outcome
`deferred_reattach_phase`.
- `heartbeat/recovery.ts`: a one-line claim hook
(`noteReattachRunsClaimed`) after restart recovery claims runs.
  - The owner of those three files agreed to these one-line gates.
- `tests/restart-chaos/daytona-image.ts`:
`PAPERCLIP_CHAOS_TEST_ACCEPT_IMAGE_CONTENT_ID` lets a local wave run a
controller-only branch against an existing Daytona image. It accepts a
content-ID mismatch only for the exact content ID of the named image,
and it prints a TEST OVERRIDE warning. This is harness-only.

## Verification

Red on master, green on this branch:

- `heartbeat/reattach-phase.test.ts` (27 tests). Five of the six
two-server scenarios were red on the earlier phase timing:
- `does not hold queued starts while the previous owner is alive and not
handing over`
- `closes the gate when the previous owner hands its runs over, and
counts reach latency from the hand-over`
- `runs the cap from the hand-over, so a wedged run ends the phase
without the overlap eating it`
- `begins when a killed owner's lease is observed expired, without any
hand-over`
- `re-arms for a later batch, including runs the previous owner started
after discovery`
- `closes the gate at once when this server claims a run between two
lease reads, dated from the hand-over`
- plus `claimQueuedRun leaves a queued run queued and unstarted, in the
admission gate's shape`, `is the same gate the assigned-work sweep and
reconciled continuation delivery read`, `ends at the cap, counted from
when the runs became claimable, so one wedged Runner cannot hold run
starts`
- `heartbeat-native-early-hand-off.test.ts`: `releases the remote
controller lease of %i live runs at once` (1, 20, 40 runs; 40 runs in 56
ms), `leaves a run another server already took over alone`, `holds the
incoming server's queued starts only from the hand-over, not while the
outgoing server still drives its runs`
- `heartbeat/run-output.test.ts`: `lets a reattach start before its
existing log is re-verified`
- `heartbeat/agent-directory-busy.test.ts`: `defers an ordinary start at
once, with no wait added to the run-start path`, `lets a reattach wait
its turn in process instead of a deferred full re-preparation`, `still
defers a reattach once its bounded wait runs out`
- `tests/restart-chaos/daytona-image.test.ts`: `accepts a mismatch only
for the exact image content ID named by the test override`
- Review fixes (red before the fix, green after):
`reattach-phase.test.ts` `starts the claim of handed-over runs within a
second of the hand-over notification`; `assigned-work-invariant.test.ts`
`takes no action while handed-over live runs are being reached, and acts
once the phase ends`; `heartbeat-native-early-hand-off.test.ts` `never
lets the outgoing server claim back a run it handed off, while the
incoming server may take it over`; `reattach-phase.test.ts` `retries a
failed first listing instead of going quiet, and still finds the
previous owner's runs`, `keeps a hand-over notification whose listing
failed until a listing succeeds`, `moves the cap earlier when a claimed
run turns out to have been handed over before its claim`;
`queue.test.ts` `returns the claim to the queue when a reattach phase
begins while the claim is being checked`; `run-completion.test.ts`
`finalizes the run log in completeRun|failRun only after a reattach's
deferred log recovery settles`. `native-session-executor.test.ts` `ends
the reattach phase once a reattached %s session is driven by this server
again, without the cap or manual settling` over codex_app_server,
opencode_server, acpx_runtime, muse_external, grok_bot_mcp and
openai_dot_mcp (6 of 6 red without the hook; the test drives the
executor with a mocked session runtime, not each real driver). Lifecycle
tests for the new worker: `does no database work and keeps no deadline
on an instance with nothing to reattach`, `waits for the hand-over
notification or the lease deadline instead of polling while the previous
owner drives its runs`, `begins at a killed owner's lease deadline,
without any hand-over or notification`, `stops with the delivery
coordinator at shutdown and ignores a later hand-over`.

Local runs on this branch: `tsc --noEmit` is clean for `server`. Passing
on this branch: `reattach-phase.test.ts`,
`heartbeat-native-early-hand-off.test.ts`, `run-output.test.ts` (53
tests together), `agent-directory-busy.test.ts`,
`heartbeat-native-restart-suspension.test.ts` (115 of 115, alone), the
new cases in `queue.test.ts` and `run-completion.test.ts`. Under load 30
to 96 on this shared machine, the full `queue.test.ts` and
`recovery-wake-herd.test.ts` runs stalled in their embedded-Postgres
setup; CI supplies them.

Daytona results that exist (local servers with Daytona sandboxes,
release-swap cells). They ran on a combined branch with this PR, #21462,
#21442 and #21362 (merged), so they cannot be credited to this PR alone:

- G5 baseline (master d1d71a7755): SIGTERM to PRP authentication 21.5 to
38 s at swap 1.
- G5c (combined branch 588c6ba267): SIGTERM to authentication 6.2 to 7.7
s at swap 1. All runs that crossed swap 2 were adopted on the same
Runner process. The staging chain (`provider recovery timed out after
900000ms`, then `resumed_not_suspended`) did not recur.
- Caveat: G5c ran before the phase timing fix in this PR. In G5c the
phase opened at process start, its 180 s cap ran out during the overlap,
and swap 2 ended `cap_reached` with 0 runs reached, 1 s before the lease
claims. The timing fix (the phase begins when the runs become claimable)
has not run on Daytona.
- G6c (12 live runs and 20 queued, two graceful swaps, no injection,
combined branch 588c6ba267): 0 failed runs at either swap; 11 of 11
survivors at both.
- Swap 1: hand-off of 11 runs in 291 ms; claims 0.3 to 2.9 s after
hand-off; authentication 7.6 to 12.5 s and identity 8.3 to 13.3 s after
SIGTERM. The 30 s target was met.
- Swap 2: hand-off took 8,562 ms and `hot_restart_prepare` timed out at
19 s, so the outgoing server drained gracefully. The incoming server did
not claim on the hand-over; it claimed only when the outgoing server
exited, 33 s after SIGTERM. Authentication was 46 to 54 s and identity
47 to 57 s after SIGTERM. The target was missed. Why the claims waited
for the exit is not explained.
- The host load was 49 to 70 from other work. 7 of 32 tasks timed out in
the queue or budget after the swaps; we suspect the overload.

## Risks

- Product trade-off: while handed-over live runs are being reached,
queued starts wait, up to the 180 s cap. On an instance with no
handed-over runs, nothing waits.
- Not proven: two-server behavior with the phase timing fix and the
notification-driven watch (no cell has run them); warm standby becoming
active; runs that the outgoing server starts after discovery.
- Early hand-off at SIGTERM orphans a server-side exec tool that is in
flight. With #21442 the next controller settles it as interrupted with
an unknown outcome; it is never run again.
- Swap 2 in G6c: the incoming server claimed only at the outgoing
server's exit. That is unexplained.
- Cross-session ownership: the gates in
`execution-recovery-resolution.ts` and
`recovery/assigned-work-sweep.ts`, and the claim hook in
`heartbeat/recovery.ts`, are in files that the restart-recovery work
owns. They are one-line gates agreed with that owner. That owner said
the gate did not need to be inside the claim transaction; that was
before the phase could begin during a claim. After review, the
in-transaction re-check in `heartbeat/queue.ts` also reads the phase,
through the existing rollback. #21427 (open) also edits
`execution-recovery-resolution.ts`, `heartbeat/queue.ts` and
`recovery/assigned-work-sweep.ts`; whichever merges second needs a
rebase. `native-runtime/native-restart-recovery.ts` (claim logic, owned
by the restart-recovery work): a 3-line guard in
`claimNativeRestartRecoveries` returns a run this process handed off at
shutdown as `blocked` / `handed_off_to_next_controller` before takeover
is evaluated, so the outgoing server never claims back a run it handed
off. #21478 (merged) added a sweep path that starts an overdue
saved-work recovery; it is deliberately not gated by the reattach phase,
because it resumes an existing run (reattach-like work), which the phase
makes room for rather than competes with.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, with tool use and code
execution in Claude Code.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

🤖 Generated with [Claude Code](https://claude.com/claude-code)

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-13 18:55:31 -06:00
Devin FoleyandPaperclip 493519fc88 fix(runner): poll a live ACPX turn without waiting out the idle read (#21525)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - The native runner (runnerd) relays provider events to the server and
waits for cumulative ACKs before it polls the provider again.
> - After a control frame, runnerd uses a 1 ms queued read only when the
executor reports a live provider poll. Otherwise it waits up to the 250
ms runtime read timeout, and any frame that arrives in that time skips
the poll.
> - Only the Codex facade reports a live poll. ACPX runs (Claude, Codex
through ACPX and others) always wait the full 250 ms after the last ACK
of every batch, and steady control traffic starves their provider polls.
> - This pull request lets an ACPX executor report a live poll while its
restored session has an active turn.
> - The benefit is that ACPX output is relayed without the 250 ms idle
wait, and tool-heavy turns no longer starve provider polls.

## Linked Issues or Issue Description

**What happened?**
A routine run started several subagents that made about 190 tool calls
in about 10 minutes. On the server, consecutive runner events arrived at
least 576 ms apart (median about 1.5 seconds), one event at a time,
while the time from each event's `emittedAt` to its insert was 0 to 0.1
seconds. The runner fell several minutes behind the provider. The
backlog then overflowed the transport's event buffer (fixed separately
in #21524) and the turn failed.

**Expected behavior**
A live ACPX turn is polled again as soon as its outstanding events are
acknowledged and no control frame is queued, the same as a Codex facade
turn.

**Steps to reproduce**
1. Run an ACPX turn that produces events while the controller sends
commands and ACKs.
2. After each ACK, runnerd calls `receive_json`, which waits up to 250
ms before it polls the provider. A frame that arrives in that time skips
the poll.

**Relevant logs or output**
```
min gap between consecutive runner events: 576 ms, median about 1.5 s, never batched
insert time - emittedAt: 0.0 to 0.1 s
```

Related: #21524 bounds the receive buffer that this backlog overflowed.

## What Changed

- `acpx_provider_backend.rs`:
`AcpxCommandExecutor::provider_poll_is_live` returns true only when the
session was restored, has no persistence error, does not wait for a
next-turn generation, and its state is `turn_starting` or `turn_active`
with an active turn, no unconfirmed provider exit, and the run it owns.
- `native_provider_backend.rs`: `SelectedExecutor` forwards
`provider_poll_is_live` to the ACPX executor.
- `tests/acpx_warm_attachment.rs`: new test through
`NativeProviderCommandExecutor`. The session is not live before
`run.prepare` or after `session.open`, is live after `turn.start`, and a
new executor that has not restored its session is not live.

What does not change:
- runnerd still polls only when the read returns no frame, so a queued
stop or suspend still runs before the next provider poll.
- The ACK gate (no poll while sent events are not acknowledged) does not
change.
- The 1 ms read is used only while the source sequence advanced since
the last quiet poll, as for Codex today, so an idle turn does not
busy-loop.

## Verification

- `cargo test --release --manifest-path runner/Cargo.toml --locked -p
paperclip-runner-core --test acpx_warm_attachment` (in
`packages/paperclip-runner`): 7 passed. The new test fails without the
new method, because the default returns false.
- `cargo fmt --all -- --check` and `cargo check --locked --workspace`
pass.
- `cargo test --release --locked -p paperclip-runner-core
--no-fail-fast`: 1,127 passed, 3 failed. The failures are
`codex_provider::tests::fenced_receipt_output_reports_capacity_without_dropping_known_frames`
("failed to signal owned process group: Operation not permitted") and
`stdio_mcp::lost_acknowledgement_blocks_calls_until_socket_replacement`
and `metadata_only_socket_changes_do_not_bypass_unknown_outcome` (relay
exit timing). They also fail without this change in the macOS sandbox
where I ran the suite, and none of them uses the ACPX executor. CI
covers them on Linux.
- The existing `durable::tests::ack_release_poll` cases cover the runner
side of the live read: an ACK releases the next poll without the read
timeout, and a command queued behind the ACK runs first.
- I did not run a live provider. After the deploy, the gap between
runner events of an ACPX run with queued output should drop below 250
ms.

## Risks

- Low risk. The change only selects the existing 1 ms queued read for
ACPX during an active turn.
- A turn that produces events continuously now polls more often. Each
poll still writes the same durable state as before, so disk writes per
event do not change.
- Analysis also found that runnerd waits for every sent event to be
acknowledged before it polls again. A bounded in-flight window would
raise throughput further, but it changes the tested rule that control
queued behind an ACK runs before the next poll. That is left out of this
pull request on purpose.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used through Claude
Code with tool use and code execution.

## Checklist

- [x] I have included a thinking path that traces from project context
to this change
- [x] I have specified the model used (with version and capability
details)
- [x] I have checked ROADMAP.md and confirmed this PR does not duplicate
planned core work
- [x] I have searched GitHub for duplicate or related PRs and linked
them above
- [x] I have either (a) linked existing issues with `Fixes: #` / `Closes
#` / `Refs #` OR (b) described the issue in-PR following the relevant
issue template
- [x] I have not referenced internal/instance-local Paperclip issues or
links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip`
URLs)
- [x] My branch name describes the change (e.g. `docs/...`, `fix/...`)
and contains no internal Paperclip ticket id or instance-derived details
- [x] I have run tests locally and they pass
- [x] I have added or updated tests where applicable
- [x] I have updated relevant documentation to reflect my changes
- [x] I have considered and documented any risks above
- [ ] All Paperclip CI gates are green
- [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups
- [x] I will address all Greptile and reviewer comments before
requesting merge

Co-authored-by: Paperclip <noreply@paperclip.ing>
2026-11-13 16:46:05 -08:00