Commit Graph
3 Commits
Author SHA1 Message Date
DottaandClaude Opus 5.5 c6c7a61e2c fix(retry): release a Retry held on a resolved decision (#21267)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - When an agent run fails, the task shows a Retry button. Admission
can hold an accepted Retry while a person must answer a pending approval
or question (`decision_pending`).
> - The hold is saved on the deferred Retry receipt. Nothing cleared
that saved reason when the person answered, so the task kept saying
"Retry is waiting for a pending approval or question" when nothing was
left to answer.
> - The deferred-receipt worker drops a refused receipt from its
schedule. No producer hinted it when the decision resolved, so the held
Retry did not start until some unrelated event rescanned the company.
> - A Retry was also held by a confirmation that was already open when
the failed turn was admitted. That turn had already run past the
confirmation, so the Retry added a gate that the original turn did not
have.
> - A person's comment on the task started a continuation, but the held
Retry receipt stayed deferred next to it.
> - This pull request makes a resolved decision release the held Retry
at once, keeps the projection from showing a resolved decision as the
wait, stops holding a Retry on a decision that predates the failed turn,
and settles a held Retry when a comment or answer continues the same
failed run.
> - The benefit is that Retry always works: a person never waits on a
decision that is already resolved.

## Linked Issues or Issue Description

**What happened?**

A board user pressed Retry on blocked tasks. No run started. The task
said "Retry is waiting for a pending approval or question to be
resolved. Your message is saved." In one task, the only interaction was
a `request_confirmation` that the previous successful run created. The
user's later comment started the turn that failed. The confirmation
stayed pending, so every Retry attached to the same held receipt. After
the user rejected the confirmation and sent a new comment, the task
still showed the same wait. No run started until a server restart
rescanned the deferred receipts about three minutes later. The task
projection said `canRetry: true` with the `decision_pending` wait.

**Expected behavior**

An accepted Retry is never held on a decision that is already resolved.
Accepting or rejecting the decision releases a held Retry right away. A
person's comment on the blocked task starts work, and the held Retry is
settled. The task projection never shows a resolved decision as the
Retry wait.

**Steps to reproduce**

1. Let an agent run on a remote runner create a `request_confirmation`
for the task.
2. Let the runner be lost so the run fails with
`native_continuation_requires_reconciliation`, and let saved work
recovery complete.
3. Press Retry. The Retry receipt is saved with `executionWait.reason =
decision_pending`.
4. Reject or accept the confirmation.
5. Observe that the receipt keeps `decision_pending`, the task keeps
showing the wait, and the Retry does not start until an unrelated
rescan.

**Paperclip version or commit**

Staging Core `57aeda190b`. Fixed on top of `origin/master` `53608bd1d6`.

**Deployment mode**

Hosted cloud (staging), remote sandbox runner.

**Agent adapter(s) involved**

`paperclip_runner` (native runtime).

## What Changed

- `hasPendingContinuationDecision` (`explicit-native-continuation.ts`)
is now the single check for a decision that holds a person's
continuation. Admission and the task projection both use it.
- For a Retry, a pending interaction that was created before the failed
source run was created does not hold the Retry. That turn was already
admitted while the decision was open. Decisions that the failed run
created (for example a plan review) still hold the Retry.
- `releaseDecisionHeldExecution` (`deferred-execution-wait.ts`) removes
the saved `decision_pending` wait from deferred receipts on the task and
hints each receipt to the deferred-receipt worker. It runs in the
resolution transaction of every interaction resolution path (through
`notifyChatConfirmationResolution`) and in question answers.
- The task projection reads the live decision state. A saved
`decision_pending` wait whose decision is resolved is not shown. This
also repairs receipts saved before this change.
- When a comment or an interaction answer continues a failed run,
`retireTaskRetriesContinuedBy` (`task-retry.ts`) settles held Retry
receipts for the same source run as `task_retry_superseded`.

## Verification

- Red/green: `cd server && npx vitest run
src/services/governed-wait-runner-loss.test.ts`. The five new tests fail
on `origin/master` sources and pass with this change (17/17 pass):
  - releases the held Retry as soon as the person chooses accept
  - releases the held Retry as soon as the person chooses reject
- never shows a resolved confirmation as the Retry's wait, and both
projections offer the same Retry
- a person's comment starts work and settles the held Retry once nothing
is left to answer
- starts a Retry at once when the confirmation was already open before
the failed turn started
- The existing tests that keep a Retry or a message on hold while a plan
review, governed tool action, secret proposal, or connection
authorization is pending still pass.
- Related suites: deferred execution wait, explicit native continuation,
interaction resolution and service, task interruption presentation and
query, task retry, deferred receipt work, interaction routes.

## Risks

- Medium-low. A Retry now starts when the only pending interaction is
older than the failed run. That turn already ran with the interaction
open, so the Retry gets no new authority. Governed tool actions still
need their own approval before the tool runs.
- Interaction resolution now also updates deferred receipts on the same
task. The update is limited to receipts with a saved `decision_pending`
wait. A receipt that is still blocked by another pending decision is
held again on its next admission.
- Issue-linked approvals (`approvals` table) do not yet hint the held
receipt when they resolve. The projection no longer shows them as the
wait after they resolve.

## Model Used

- Claude Opus 5.5 (`claude-opus-5-5`), Anthropic, used through Claude
Code with tool use (shell, file edit, browser for read-only staging API
reads).

## 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>
2026-11-12 22:58:48 -06:00
DottaandPaperclip c32ecf6657 fix(server): schedule only outstanding merge confirmations (#18382)
Reviewed and verified at bc22051c56.

Co-Authored-By: Paperclip <noreply@paperclip.ing>
2026-10-31 13:44:46 -05:00
DottaandPaperclip ee67d2ece9 fix(server): wake terminal confirmation recovery from committed changes (#17651)
<!-- entire-trail-link-start -->
https://entire.com/gh/paperclipai/-/trails/202/changes/paperclip/202
<!-- entire-trail-link-end -->

## Thinking Path

> - Paperclip persists confirmation clicks and interaction decisions
before dispatch completes.
> - A crash after a decision commits can leave a processing confirmation
action.
> - Periodic publication maintenance repeatedly scans those actions even
when idle.
> - Saved decisions and the supported board/admin writers already
identify when settlement can change.
> - This owner reconstructs durable actions, reconciles real changes and
then becomes quiet.

## Linked Issues or Issue Description

This existing terminal-confirmation slice follows merged #17649. It
removes only terminal-confirmation recovery from periodic chat
maintenance while retaining the explicit manual API and existing durable
action/interaction/publication rows.

**Problem:** An accepted response can commit before its dispatch
settles. Removing periodic recovery without covering board decisions,
expiry and ownership changes would strand processing actions. Pending
human decisions must remain recoverable without polling indefinitely.

**Expected behavior:** Committed supported changes wake the scoped
owner. It reconciles terminal responses once, parks unresolved human
decisions, preserves concurrent commits/restart/drain and becomes quiet
when settled.

## What Changed

- Register terminal-confirmation work with the existing delivery
coordinator and scoped lifecycle. Reconstruct processing action IDs,
bound selection pages, preserve concurrent wakes and keep current
authorization in the existing interaction owner.
- Add the existing transaction-aware confirmation notification topic at
real resolution writes and callback settlement. Reuse milestone signals
instead of adding per-caller recovery wrappers.
- Cover ordinary and checkbox confirmations, suggestions and connection
intents through their owning service. Cover board accept/reject,
cancellation/withdrawal, opt-in comment supersession, expiry,
reassignment and already-satisfied connection resolution.
- Preserve task/actor authority, one-use callback tokens and immutable
provider publication semantics. Explicitly fail malformed missing
bindings instead of silently retaining unrecoverable processing actions.
- Remove only this duty from recurring publication maintenance. Add no
table or scheduler infrastructure. A concurrent partial index on
existing action rows prevents recovery discovery from scanning completed
history.

## Verification

Published repair head: `a35e78d076c1b618bc643391b7520fa6952fc996`,
targeting master after normal integration of
`644937c93150c423f12674992c7ea1e8b5dbc98c`. [Detailed acceptance and
preserved
failures](https://github.com/paperclipai/paperclip/blob/a35e78d076c1b618bc643391b7520fa6952fc996/doc/plans/2026-10-25-terminal-confirmation-sleep-audit.md#bound-retained-history-reads--october-28-review-repair).

- **New query-cost regression:** Capture the actual SQL emitted by
startup and company-scoped confirmation discovery. Seed 20,000
completed/unrelated action rows and one pending human decision. Before
the index, the regression fails because recovery examines 20,009 action
rows. After the partial index, startup and scoped discovery must examine
at most two action rows and fewer than 20 table/index buffers; the
no-pending scoped path must also stay below 20 buffers. Planner settings
are unchanged, and returned IDs remain correct.
- Match startup keyset ordering to `(company_id, id)`, the partial index
order. The index alone still let PostgreSQL choose the primary key and
scan 20,206 rows for `ORDER BY id`; the final query fixes this without
forcing a planner setting or adding a second index. Fixed kind/status
predicates stay literal for prepared-plan index eligibility.
- Add the generated `0398_yielding_random.sql` migration, using `CREATE
INDEX CONCURRENTLY IF NOT EXISTS` and the existing top-level migration
runner. Keep the five newest generated snapshots and preserve every
prior SQL/journal entry. No work table is introduced.
- **110 distinct cases pass:** 17 final confirmation
lifecycle/query-cost cases (1,671 skipped), seven real Runner descriptor
variants (295 skipped), and all 86
app/startup/bundled-plugin/plugin-database cases. Server/database
typechecking and builds, migration safety/numbering, module boundaries
and a fresh PostgreSQL migration through 0398 pass. The final test-only
vacuum precondition is followed by another 17-case confirmation pass and
server typecheck. Counts are not added across repeated runs. Normal
integration of master `644937c93150c423f12674992c7ea1e8b5dbc98c`
preserves the complete owned patch and all 18 unrelated master paths.
- Previous landed-parent acceptance: 29 distinct cases (16 confirmation
lifecycle/producer cases with 1,671 skipped, plus all 13 startup cases)
passed on `32e64a7e9ea5b457e54ca7b268be90e1e155b451`; the earlier
80-case matrix remains scoped to
`c1e8542b588a4fa59b62f9829541d3daea4f1ddc`. Counts are not added across
repeated runs.
- Actual producer cases prove parking/quiet, caller-owned outer
transaction rollback, commit-driven automatic settlement, restart/drain
and concurrent commits. Checkbox, suggestion, connection-intent and
Telegram/Slack callback paths retain existing authority and immutable
publication semantics.
- **Fresh exact-head CI and Greptile 5/5 with no unresolved actionable
findings are required.** The preceding head's Runner CI failed its
timeout-attachment durable-suspension case (1 failed/1,955 passed/4
skipped); that failure is retained and not counted as a pass. The new
head must pass this job. No full local repository, live provider or
physical suspension claim is made.

The owner audit retains the earlier missing-board-producer regressions
and their corrections, including the invalid supersession fixture that
assumed a new checkbox expires an older one. Existing behavior was
preserved by using the supported opt-in human-comment path. Prior
scheduler/backoff, route and plugin test failures remain at their
original revisions; no old pass is substituted for current remote gates.

## Risks

- Process-local notifications cover supported application writers and
their shared pools, not independent SQL writers or other replicas.
- Processing actions waiting on a human response are deliberately
parked. Supported interaction resolution/expiry writers must use the
notification boundary; a missing producer delays settlement until
restart.
- Invalid bindings have explicit failure evidence. Source, actor, task
and publication authority remain checked by their existing owners.
- Following continuation/GitHub duties remain separate. Cloud sleep
gates and external alarms are unchanged; local worker quiet does not
establish physical host/database suspension or fleet savings.

## Model Used

OpenAI GPT-6 / Codex, with reasoning, repository inspection, code
execution, and PostgreSQL tests. The exact runtime model identifier and
context-window size are not exposed in this session.

## 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>
2026-10-28 15:00:31 -05:00