mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-11 23:36:51 +02:00
codex/plugin-task-execution
3
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 DottaandPaperclip
|
5842185e4f |
fix: surface native cleanup quarantine and add chat qualification evals (#13775)
## Thinking Path > - Paperclip lets people manage AI agents and their work. > - Agent Chat needs reliable native execution before native runners become the onboarding default. > - Existing stories covered idle reassignment and controller restart, but not an executing worker handoff or worker process loss. > - Status answer tests also need to reject stale claims and invented facts. > - This pull request adds six opt-in full-stack cells with independent state assertions and retained evidence. > - The probes exposed a misleading Retry across server projection and recovery-banner paths; the fix reports the blocked recovery honestly. > - The tests preserve failures without changing recovery policy, production prompts, or onboarding defaults. ## Linked Issues or Issue Description Refs: #13762. Related: #13765 (Retry targets the latest failed attempt), #13753 (task context ownership), #13746 (native recovery work). ## What Changed - Add active reassignment with saved draft and plan preservation, old-worker cancellation, and successor completion checks. - Preserve recovery-needed projection when native cleanup fails before its coordinator exists, refuse a generic retry that would immediately fail again, and replace the recovery banner's misleading Retry with Inspect run. - Add verified local worker process loss with a required successful continuation; retain a failing qualification result when recovery is unavailable, while independently verifying the UI/API refuse doomed retries. - Add two-turn factual answer checks for current blockers, stale claims, inactive backlog work, and unknown facts. Retain prose for separate semantic review. - Add positive and negative oracle calibration and document fault isolation, cleanup, billing, and qualification limits. ## Verification - Eval TypeScript check passes. - All 442 eval support tests pass locally. The 89 focused server tests and server typecheck pass. Six recovery-banner UI tests and token gates pass. - Initial new-cell campaign: https://github.com/paperclipai/paperclip/actions/runs/35657128077. All six results are retained; four failed on fixture-contract issues and two exposed real worker cleanup quarantine. - All 26 existing native onboarding cells: https://github.com/paperclipai/paperclip/actions/runs/35656761484 (26/26 passed on master |
||
|
DottaandPaperclip
|
0103cb2292 |
test: use neutral references in chat hiring evals (#13752)
## Thinking Path > - Paperclip helps people manage agents and their work. > - The agent chat hiring eval checks delegation, saved output, and worker reuse. > - It asks a worker to include a unique reference in each document. > - The open wording let the worker choose a credential-style label. > - The existing redactor removed that reference and failed the coordination check. > - This PR specifies a neutral reference line while keeping the same grading checks. ## Linked Issues or Issue Description Refs #13741. **What happened?** The Codex hiring eval saved a checklist with `Tracking token: [REDACTED]`. The runner treats the chosen label as credential syntax. The previous prompt only asked for the identifier and did not choose its label. This tests a redaction boundary unrelated to hiring and reuse. **Expected behavior** The coordination fixture requests ordinary business content with a neutral reference label. The grader still requires the exact identifier in the saved output. **Steps to reproduce** Run `agent-chat-hardening.runner-codex.local.hire-delegate-reuse` with the previous fixture. The retained failing attempt is in [campaign 35617045456](https://github.com/paperclipai/paperclip/actions/runs/35617045456). ## What Changed - Request `Reference: ...` in the checklist and review documents. - Increment the hardening suite definition to version 5 and record the reference format. - Document the fixture boundary. Production redaction and all grading checks stay unchanged. ## Verification - `pnpm test:e2e:runner:typecheck` passed. - `pnpm test:e2e:runner:unit` passed: 424 tests in 37 files. - The catalog lists exactly the Codex and Claude local hiring cells for the selected case. - [Live campaign 35620731321](https://github.com/paperclipai/paperclip/actions/runs/35620731321) passed both selected local cells on `d8d7afe21`: native Codex (`gpt-5.6-sol`) and Claude (`claude-sonnet-5`). Both passed on attempt 1, including cleanup. [Published eval report](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-35620731321-1/). - Inspected saved evidence: each provider preserved both exact reference lines, used the same hired worker for checklist and review, and completed the final status query. Definition version 5 and hash `2c1c9f754c049c3e7cafd6c8d0a137e7ca06d546ce1e7b07ad5b7823f0f28db9` distinguish these results from the prior prompt. - [PR CI 35620741101](https://github.com/paperclipai/paperclip/actions/runs/35620741101): full build, type checks, and test partitions passed. One unrelated Telegram integration test initially failed because two random fixture company IDs produced the same seven-character issue prefix. The failed shard passed on one retry; no test code was changed. All latest-head merge checks are green. - Greptile reviewed `d8d7afe21` at 5/5 with no review threads. ## Risks The live models can still fail the coordination workflow. This change does not qualify or change credential-redaction policy. Earlier failed attempts remain part of the evidence; the new fixture has a distinct definition version. ## Model Used OpenAI Codex, GPT-6, with repository tools and code execution. The exact deployment model ID and context window are not exposed in this task. ## 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 Co-authored-by: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
9d19f98b50 |
fix: harden native chat recovery and add coordination evals (#13741)
## Thinking Path > - Paperclip helps people manage AI agents and their work. > - Agent chat uses native runner sessions to plan, delegate, and track that work. > - A user can press Stop while the native session is still starting. > - The server can acknowledge that Stop without dispatching it, then let the session submit a turn. > - This leaves chat recovery waiting for an execution that the user expected to stop. > - This PR waits for the startup handle, dispatches cancellation, and prevents a late startup from submitting a turn. > - New full-stack evals check the resulting records and outputs across Claude and Codex. > - Those evals also exposed missing ACPX readiness fields, unbounded polling, and an old-run identity check that rejected valid warm handoffs. ## Linked Issues or Issue Description **What happened?** Stop during native startup could record an acknowledged cancellation with `dispatched: false`. The provider could then begin work. A subsequent `/new` stayed queued. A remote Claude follow-up also exhausted the command journal while probing warm-session readiness: ACPX never returned the readiness fields required by the shared transport. Once readiness worked, attachment incorrectly compared the next run descriptor against the old run ID. The 25 ms polling loop could issue 4,800 commands during its two-minute wait, beyond the 500-command bound. The existing chat eval treated lifecycle logs as proof of an active provider turn, so it did not distinguish startup cancellation from active-turn cancellation. **Expected behavior** A Stop during startup must reach the pending session. A late session must not submit a prompt after Stop. Recovery must retain control when startup exceeds the bounded wait. Chat evals must check saved task state, document contents, worker identity, account binding, and duplicate effects. **Steps to reproduce** 1. Start a native Claude or Codex chat turn. 2. Press Stop after process startup is requested but before the provider turn starts. 3. Send `/new`, then send a fresh message. 4. On the affected base, cancellation can be acknowledged without dispatch and the reset stays queued. **Paperclip version or commit** The live Claude baseline reproduced this on `29d6b3509`. The branch also includes master commit `0f5fafe16`. Related work: #13678, #13686, #13693, #13291, #13738. A separate runner reliability branch also contains a startup-wait fix. Its overlap must be reconciled before merging; this branch additionally prevents prompt submission after a late startup. ## What Changed - Wait for a pending native startup before acknowledging a run-scoped Stop. Preserve the existing recovery error when that wait expires. - Keep a Stop guard on startup. Cancel a late handle before it can submit a provider turn. - Add regression tests for normal handle publication and publication after the Stop deadline. - Back off blocked warm-attachment probes. Keep the fast two-snapshot barrier, fail closed, and record changed blockers. - Add red/green tests for delayed readiness, persistent blockers, alternating readiness, and readiness near the deadline. - Publish ACPX readiness and blockers. Preserve the old authority’s event acknowledgement barrier; only settled sessions can proceed to attachment. - Bind warm ACPX descriptors to the validated next authority while retaining old-run event correlation until activation. Preserve session identity and provider profile checks. - Exercise two consecutive run rotations through a qualified fake sidecar, verifying checkpointing, provider identity, pre-activation rejection, and new-run work admission. - Separate startup and active-turn cancellation checkpoints in the browser eval. - Add 18 explicit native chat eval cells: 12 local and 6 Daytona cells across Claude and Codex. - Cover hiring and reuse through managed AI accounts, source-based review, current blocked-task status, request replay after a lost HTTP acknowledgement, server restart continuity, and Stop/reset continuity. - Use ordinary production agent instructions. Enable API tools only for the two coordination cases that need them. - Calibrate the matchers with invalid records and outputs. Require remembered context after restart and a structured status snapshot that distinguishes the current blocker from history and task status from active execution. Compare the public issue mutation contract and relationships during read-only reporting. Preserve before/after source records in failed eval evidence. - Fix the lost-ack browser harness and verify it against a real HTTP server. Check the chat composer after restart instead of waiting for an unrelated document lifecycle event. - Document the scope and limits of each case. ## Verification - The startup regression failed on the unfixed executor and passed after the fix. - `pnpm test:e2e:runner:typecheck` passed. - `pnpm test:e2e:runner:unit` passed: 424 tests in 37 files. - `pnpm exec vitest run server/src/services/native-runtime/native-session-executor.test.ts` passed: 385 tests. - [Baseline live campaign](https://github.com/paperclipai/paperclip/actions/runs/35608208868): Claude Stop reproduced the bug. Codex Stop and Claude hire/reuse passed. Codex delegation was blocked by provider capacity. - [Eval-only startup campaign](https://github.com/paperclipai/paperclip/actions/runs/35609479786): both providers failed as expected. Both persisted `dispatched: false` and left `/new` queued. - [First fixed campaign](https://github.com/paperclipai/paperclip/actions/runs/35610533706) on `c9e95797d`: 10/18 cells passed. Startup Stop passed for both providers. Failed cases exposed eval harness defects and remote continuity failures. All attempts remain available. - [Original workflows and stronger memory checks](https://github.com/paperclipai/paperclip/actions/runs/35611896649) on `c04324fab`: 9/12 passed. Reassignment, local restart memory, and startup Stop passed for both providers; Codex remote restart passed. Claude remote restart exposed the missing readiness contract. Two Codex planning cells hit provider capacity. - [Unchanged-model retry](https://github.com/paperclipai/paperclip/actions/runs/35613854548): Codex planning and backlog creation both passed. - [18-cell campaign with ACPX readiness](https://github.com/paperclipai/paperclip/actions/runs/35614586963) on `6a98ef743`: 16/18 passed, including all local/remote Stop and committed-send cases. Claude remote continuity exposed the next-authority check, now fixed. Codex hiring produced its checklist, but the runner redacted the requested marker after it appeared as “Tracking token: …”. That content-redaction policy is unchanged and remains an explicit limitation. - [Structured status grading](https://github.com/paperclipai/paperclip/actions/runs/35614954725) on `50448c228`: both providers passed on their first attempt, including cleanup. - [Complete read-only state grading](https://github.com/paperclipai/paperclip/actions/runs/35616089011) on `551e13892`: both providers passed. - [Final ACPX handoff and hiring retry](https://github.com/paperclipai/paperclip/actions/runs/35617045456) on `cbd637587`: all three Claude Daytona cases passed (restart continuity, active Stop/reset, and lost-ack replay). Codex hiring reproduced the content-redaction failure: the saved checklist contained `Tracking token: [REDACTED]` instead of the required business marker. All four cases completed cleanup successfully. [Published report](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-35617045456-1/). The only subsequent commit adds the qualified-sidecar integration test; production code is identical to this live proof. - `pnpm test:e2e:runner:browser-support` passed: 5 browser tests without paid models. - Runner TypeScript typecheck passed. All 5 warm-readiness tests pass; two failed with the prior fixed-rate loop, and the late-readiness test failed before the pacing correction. - ACPX readiness and warm-identity regressions each failed before their fixes. All 292 runner-core Rust library tests passed. The qualified-sidecar integration test passes. Rust formatting is checked. - Status-grader regressions for misleading historical mentions and previously unchecked mutations each failed before tightening the oracle and pass now. - [Latest-head CI](https://github.com/paperclipai/paperclip/actions/runs/35617522307) passed on `a4093c8f1`: full build, type checks, test partitions, browser E2E, and native runner checks. Two unrelated tests initially failed (Sentry fixture release attribution and local-service fixture readiness); both passed locally together (35 passed, 5 optional SDK tests skipped) and on the failed-job retry. No changes were made to those tests. - Greptile reviewed `a4093c8f1` at 5/5; both earlier findings are fixed and all review threads are resolved. - The paid live suite is not fully green: the reproducible content-redaction case remains red. This is separate from the passing PR merge checks. No production content-redaction, prompt, model, or completion-policy change is included. - Managed-account hiring and review cases explicitly enable API tools; these do not qualify default new-user onboarding. ## Risks - Stop can wait up to 30 seconds for startup, then use the existing pending-recovery path. This does not prove that remote cleanup has finished. - Blocked warm readiness adds up to 750 ms between later probes with the two-minute remote budget, or about 32 ms with the default five-second budget. Ready sessions retain the short second barrier. - Paid evals can fail because of provider capacity or agent decisions. Each failure needs evidence-based classification. - The HTTP request replay case checks comment idempotency and duplicate effects. It does not prove replay safety for an ambiguous provider tool call. - The new suite is opt-in. It does not increase the default paid campaign. - No production prompts or model selection change. Review-handoff behavior and content-redaction policy remain separate product decisions. The latter can remove harmless business content that looks like credential syntax; the failing attempt is retained. ## Model Used OpenAI Codex, GPT-6, with repository tools and code execution. The exact deployment model ID and context window are not exposed in this task. ## 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 --------- Co-authored-by: Paperclip <noreply@paperclip.ing> |