Commit Graph
3 Commits
Author SHA1 Message Date
DottaandClaude Opus 5.5 d4498ae7ac fix(runner): settle every recorded tool result and free sessions blocked by a closed turn's pending call (#21490)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Agents call Paperclip tools through the Runner. The Runner records
each call and returns the controller's result to the provider (here
OpenCode)
> - In chat-eval cohort 13, a `list_workflows` result was recorded but
never reached OpenCode. OpenCode waited about 290 s, and the call stayed
pending
> - The next run's `run.attach` then failed with
`durable_pending_tool_calls`, so the person's follow-up question never
got an answer
> - This pull request fixes the result shape that caused the rejection,
makes the Runner settle every recorded result, and lets the next run
settle a call that a closed turn left pending
> - The benefit is that a tool result always reaches the agent, and a
stuck call can no longer block every later turn

## Linked Issues or Issue Description

**What happened?**
Chat-eval cohort 13 (OpenCode,
`openrouter/deepseek/deepseek-v4.1-flash-1017`, master `df1b12ad8e`). In
`cases.case-history` and `cases.create-case`, the CEO called
`list_documents` (or `list_projects`) and `list_workflows` (call id 5)
together. The server answered `GET /companies/{id}/workflows` with 200
in 584 ms. The OpenCode MCP bridge diagnostic for call 5 says `outcome:
"unanswered"`, `written: false`, `handlerMs: 289038` (290723 in
create-case). The sibling call was answered in about 1 s. The run ended
with `runner_settlement_incomplete`: `semanticTools.pending` held call 5
and its `semantic_tool.result` delivery command had status `failed`. The
next question started a run whose `run.attach` was refused: `run.attach
requires a settled Codex provider session: durable_pending_tool_calls`.
The run failed in 17 s, and the question was not answered.

Cause: `executeNativeWorkflowTool` returned the Runner API result
unchanged: `{ ok: true, status: 200, operationId: "GET
/api/companies/{companyId}/workflows", ... }`. Every other API-backed
tool renames that field to `apiOperationId`. The Rust tool bridge
(`semantic_response_value`) reads an object with a boolean `ok` and an
`operationId` as a semantic envelope that must name this tool and call.
The REST route did not match, so `apply_result` failed with `semantic
result envelope does not match its provider call`. Nothing was sent to
the provider, and the call stayed pending forever. The red test
reproduces this exact error. The parallel call is not the trigger: the
sibling call settled normally, and the OpenCode MCP bridge answered
other parallel pairs in the same runs. `list_workflows` first appears in
cohort 13, so cohorts 11 and 12 did not hit it. All four workflow tools
had this shape.

**Expected behavior**
Every recorded tool result reaches the provider. When a recorded result
cannot be delivered, the provider gets an error at once and the call
settles. A call that a closed turn left pending does not block the next
run.

**Steps to reproduce**
1. Run an OpenCode agent with Cases enabled on the durable Runner.
2. Ask a question that makes the agent call `list_workflows`.
3. See the call wait about 290 s. Then ask a follow-up question and see
`run.attach` fail with `durable_pending_tool_calls`.

**Paperclip version or commit**
`master` at `df1b12ad8e`.

## What Changed

- `server/src/services/native-runtime/native-workflow-tools.ts`:
workflow tool results name the route `apiOperationId`, like case tools
and `call_api`.
- `provider_bridge.rs`: add `settle_undeliverable_result`. When the
bridge rejects a recorded result for a call it still holds, the call
settles with a bounded, non-retryable runner error. Conflicting or
unknown results still fail as before.
- `provider_backend.rs` (`deliver_semantic_result`): use that
settlement, deliver the error to the provider, and record a
`semantic_tool_result_undeliverable` diagnostic with the call,
operation, reason and the rejected result's digest.
- `provider_bridge.rs` and `provider_backend.rs` (`attach_run`): add
`expire_closed_turn_calls`. When nothing is live (no active or ambiguous
turn, no provider request waiting), `run.attach` settles calls that a
closed turn left pending as an expired failure with an unknown outcome,
and records `semantic_tool_result_expired`. The provider stopped waiting
when its turn ended, and the new run's controller never delivers the old
run's results. A call that a live turn still waits on still blocks
attachment. This follows the existing `tool_call_interrupted_by_restart`
rule: never execute the operation again, and record an unknown outcome
when no result can ever arrive.
- Docs: `doc/run-log-events.md` and
`packages/paperclip-runner/docs/durable-recovery.md` describe both
diagnostics.

## Verification

- Red/green, Rust (`cd packages/paperclip-runner/runner && cargo test -p
paperclip-runner-core --test codex_provider --
a_rejected_recorded_result run_attach_expires`):
- Before:
`a_rejected_recorded_result_still_answers_the_provider_and_frees_the_session`
fails with `semantic result envelope does not match its provider call`
(the production error).
`run_attach_expires_a_call_left_pending_by_a_closed_turn` fails with
`run.attach requires a settled Codex provider session:
durable_pending_tool_calls` (the production refusal).
- After: both pass. The fake provider logs `tool-response:failure`, so
it got an answer at once. The next `run.attach` succeeds.
- Parallel calls (`cargo test -p paperclip-runner-core --lib --
provider_bridge`):
`a_rejected_result_in_a_parallel_pair_still_settles_its_own_call` sends
the cohort 13 pair (`list_documents` + `list_workflows`). The rejected
call settles, its sibling keeps its own result, and the next turn
starts. `only_a_closed_turn_expires_its_pending_calls` checks that an
open turn expires nothing.
- Upgrade and mid-run cutover:
`run_attach_expires_a_call_left_pending_by_a_closed_turn` saves state
with a pending call after the turn ends, stops that Runner, and attaches
with a new Runner process from the saved state.
`durable_backend_settles_pending_tools_when_recovery_finds_the_turn_ended`
(unchanged) still passes: a late result in the same run still settles as
`settled_after_turn`.
- Red/green, server (`cd server && npx vitest run
src/services/native-runtime/native-case-tools.integration.test.ts -t
workflows`): before, `expected { ok: true, status: 201, … } to not have
property "operationId"`; after, 2 passed.
- `cargo test -p paperclip-runner-core`: 643 lib tests pass. One lib
test (`fenced_receipt_output_keeps_prefix_when_stdout_never_closes`)
failed once with `EPERM` while signaling a process group under parallel
load. It passes alone 3 of 3 and does not touch this change. The
integration test binaries pass.
- Paid narrow run for `cases.case-history` and `cases.create-case`: see
the comment below.

## Risks

- Previous-release story: no persisted shape changes. State saved by the
previous Runner with a call pending after its turn closed carries
forward. The new Runner settles that call at the next `run.attach`, and
the session continues (tested from saved state with a new process). A
previous-release Runner that receives the corrected `apiOperationId`
result delivers it normally, because the result is no longer read as an
envelope.
- Behavior change: a call left by a closed turn is now settled at the
next run's attachment, not kept forever. Its outcome is recorded as
unknown, and the operation is never run again. A provider turn that is
still live keeps the old refusal (`durable_pending_tool_calls`,
`provider_side_effect_pending`), and the controller still never replaces
the session.
- A recorded result that the bridge rejects now becomes an error for the
agent, not an endless wait. The agent is told the action may have
completed and must check the current state before repeating it.
- Latency: nothing is added before turn start. The attach-time check
walks the in-memory pending calls (usually none) and adds no I/O beyond
the existing state save. The rejected-delivery path now answers in the
same command instead of after the provider's ~290 s client timeout, and
the follow-up run no longer fails after 17 s.

## Model Used

- Claude (Anthropic), model ID `claude-opus-5-5`, 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

Refs #21398 (first release where `list_workflows` was called in this
eval), #19733 (the attach refusal), #20886 (earlier OpenCode delivery
fix).

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

---------

Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>
2026-11-14 14:24:21 -06:00
DottaandClaude Opus 5.5 eb5ad4c920 fix(cases): say a Cases-disabled denial is final and name the Experimental setting (#20892)
## Thinking Path

> - Paperclip is the open source app people use to manage AI agents for
work
> - Cases is an experimental feature; an instance admin turns it on in
Experimental settings
> - When Cases is off, every case and workflow route answers 403 "Cases
are disabled" with no other information
> - A chat agent asked to "open a customer case" could not tell whether
the denial was final or what to offer
> - In one eval run it created a project and a task in place of the
case; in another it spent most of its turn searching routes before it
replied
> - This pull request gives the denial a code and a final, actionable
remediation, like the board-only denial
> - The benefit is a fast, truthful reply that names the setting and
lets the person choose

## Linked Issues or Issue Description

**What happened**
In the chat eval cases.cases-disabled ("Open a customer case for this
bug." with Cases off), the agent got 403 "Cases are disabled". In one
run it then created a project and a bug task without asking. In a later
run it searched many routes and instance settings for about 100 seconds
before it replied.

**Expected behavior**
The denial says that Cases is off, that the denial is final, where an
instance admin turns it on, and that the agent must not create a
substitute record before the person chooses. The agent replies at once
and offers the setting or an ordinary task.

**Steps to reproduce**
1. Turn off Cases in Experimental settings.
2. As an agent, call GET /api/companies/{companyId}/cases.
3. See a bare 403 "Cases are disabled" with no code or remediation.

## What Changed

- Rebased on #20894, which already gave the case routes a
`cases_disabled` code, a final remediation and the hidden-settings
check. This PR moves that remediation into one helper used by every case
and workflow path, and adds: no substitute task or project before the
person chooses, and the offer of an ordinary task.
- New `casesDisabledError()` in `server/src/services/cases-disabled.ts`:
same 403 message, plus `code: "cases_disabled"` and a remediation. The
remediation says the denial is final, forbids a substitute task or
project before the person chooses, links `[Experimental
settings](/<prefix>/company/settings/instance/experimental)`, and offers
an ordinary task.
- Native tool results carry only the error message, so the native case
tools, workflow tools and runner tool authority inline the remediation
in the message (`casesDisabledError({ inlineRemediation: true })`).
- Case routes, workflow service, case handoffs, native case tools,
native workflow tools and the runner tool authority use the helper
instead of a bare `forbidden("Cases are disabled")`.
- Tests: `cases-routes.test.ts` checks the code and remediation;
`native-case-tools.integration.test.ts` checks the native message
carries the remediation (red on master, green here).

## Verification

- `npx vitest run src/__tests__/cases-routes.test.ts
src/services/case-handoffs.test.ts
src/services/native-runtime/native-case-tools.integration.test.ts
src/__tests__/heartbeat-case-handoff.test.ts` (166 passed).
- The cases.cases-disabled chat eval is rerun on a temporary branch;
results are recorded in the PR conversation.

## Risks

- Low risk. The status and message do not change. The error body adds
`code` and `remediation`.

## Model Used

- Anthropic Claude Opus 5.5 (`claude-opus-5-5`), 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>
2026-11-11 13:36:08 -06:00
DottaandPaperclip 0a5f1546ee feat(cases): add reusable prompt workflows (#17114)
## Thinking Path

> - Paperclip manages AI agents and their work.
> - Cases keep facts, documents, and one primary task across agent
handoffs.
> - Each case currently needs its own procedure text.
> - Agents need to reuse procedures and move cases between them without
replacing the task.
> - This change adds company workflow definitions with a name and
Markdown prompt.
> - Existing handoff and document transactions preserve authority,
revisions, and recovery.
> - Cases remain behind `experimental.enableCases`.

## Linked Issues or Issue Description

Refs #16959. Related proposal: #3951 uses task graphs and a visual
editor. This change extends existing Cases with prompt definitions and
keeps its current handoff model.

**Problem or motivation**

Agents cannot share a reusable procedure across cases or inspect a
case's movement between procedures.

**Proposed solution**

Add name-and-prompt workflow definitions. Adopt a prompt into the
primary task's revisioned workflow document through a handoff. Keep
phases as free text.

**Alternatives considered**

Defer graphs, workflow versions, run entities, role schemas, authoring
screens, and Spec Kit integration.

**Roadmap alignment**

This extends the existing Cases experiment and same-task review flow. It
adds no work queue or new execution engine. The maintainer approved this
scope in the implementation plan.

## What Changed

- Add company-scoped definition storage, shared request schemas, REST
routes, and four optional Cases tools.
- Accept a workflow ID at kickoff. Agent handoffs can adopt, refresh,
preserve, or detach a workflow.
- Capture prompt text and the document base revision at acceptance.
Apply workflow, phase, and document revision together after outgoing
execution settles.
- Hold conflicting document edits for recovery. Preserve cancellation,
replay, and restart behavior.
- Show current workflow names and captured transition names/phases in
Work, history, and Audit.
- Update instructions, API/tool contracts, capability maps, and the
isolated workflow journey.
- Pin bounded workflow prompt reads to a content fingerprint;
continuation conflicts on edits and can restart safely.
- Remove experimental Cases/workflow/Mermaid guidance from the always-on
native prompt. Extract runtime-context prompt strings into a
CODEOWNERS-protected constants file while preserving the public imports.
- Refresh Copilot’s pinned source identity to v57, including the
extracted prompt file and retaining historical identities.

## Verification

Local checks: repository `pnpm -r typecheck` and `pnpm build` pass on
the prompt-extraction tree (`88222d45d8`). Full repository coverage
previously completed through the supported isolated shard groups (1,082
general server suites, 161 serialized suites, and all three workspace
groups). The earlier default serial run was interrupted and remains
incomplete. The first sharded run retained the failures described below;
targeted corrections and separate limited-concurrency reruns pass. All
CI jobs, security checks, and the 5/5 Greptile review passed on
`fc1bd79c28`. Fresh CI and review are pending for the prompt extraction
in `88222d45d8`. Focused verification passes: real-database handoff
recovery and reuse; 2,002 native-tool/API contract checks; 144 Runner
catalog/discovery checks; 27 Runner context/coverage checks; 37
case/Audit UI checks; and Product E2E oracle calibration. Token gates,
migration safety/numbering, module boundaries, capability maps, and
generated tool/Storybook catalogs pass. The final route/context
corrections pass 49 API/native-tool tests, 90 real-database handoff
tests, and server typecheck; 142 Copilot installation/sidecar tests pass
with the refreshed identity.

Greptile’s first review found that successive prompt pages could mix
definitions after an edit. The fix returns a content fingerprint,
requires it for continuation, and rejects edits with 409. Strict
live/catalog schemas pass (171 Runner checks). The 2,005 API/OpenAPI
tests pass; the initial native regression caught compression’s weak ETag
prefix, and all 23 native integration tests pass after handling it. The
regression covers same-length edits and a four-byte restart window. No
stored version fields or entities were added.

The prompt extraction passes all 462 checks across six focused suites,
including persisted-context recovery and provider transport. The first
local transport attempt lacked the required debug Runner binaries (102
failures, retained); after building the normal prerequisites, all 283
tests in that suite pass without source or assertion changes. Existing
saved snapshots retain their recorded prompt text, and new executions
use core-only prompt v7. The extracted prompt source is now covered by
both CODEOWNERS and Copilot’s source fingerprint.

Live-provider evidence is separate from those simulated checks: the
isolated API-key campaign `reusable-workflows-api-final-2026-10-21`
passed with local code matching `a1d73b4652`. Four actual Codex API runs
read the adopted task documents, moved one case A → B → A on the same
task, completed normally, and passed all nine durable checks. Browser
assertions verified the current workflow/phase, historical names/phases,
five handoff records including terminal completion, and the completed
original task. Cleanup passed. The local report remains under
`tests/runner-e2e/results/reusable-workflows-api-final-2026-10-21/`; no
credentials or screenshots are committed.

CI initially caught an omitted route-inventory entry and a stale pinned
instruction-source hash; both are corrected and separately verified.
Local serialized coverage also recorded an unrelated environment-route
socket disconnect and activity-route timeout; their separate 11-test and
17-test suites passed without code or timeout changes. An unchanged
chunked-webhook check also passed its separate four-test rerun. The
initial adapter/database workspace groups hit time and resource limits
under default worker counts; complete reruns with `VITEST_MAX_WORKERS=2`
pass, including all 55 database files and 211 tests. The UI/CLI groups
passed 837 files and 9,442 tests in the original run. Original failures
remain retained and are not regraded.

Earlier failed live campaigns remain unchanged: subscription HTTP 429
before agent execution; an obsolete exact-four-row oracle that rejected
the valid terminal fifth row; and a chat-marker expectation that did not
account for the existing explicit-comment presentation policy. The final
fixture requires terminal completion and places the marker in its
authoritative comment. Positive/negative oracle tests and the fresh live
campaign pass.

## Risks

- The additive migration adds a table and nullable case/handoff fields.
Existing rows keep null workflow metadata; no assignments are inferred.
- Adoption can wait on an intervening task-document edit. Recovery must
cancel the old intent and accept a new one after readback.
- Shared definition edits affect future adoptions. Existing case copies
keep their revisions.
- The feature remains experimental. This PR does not merge or deploy the
change.
- Live acceptance covers native Codex before the prompt extraction; this
cleanup has contract/transport verification rather than a new
live-provider campaign. The refreshed Copilot identity has static
compatibility evidence, not new live-provider qualification.

## Model Used

OpenAI Codex, configured model ID `gpt-6-astra`, xhigh reasoning. Used
terminal execution, file editing, and tool calls. The active
context-window size is not exposed.

## 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>
2026-10-21 15:00:12 -05:00