2 Commits

Include renames

Author	SHA1	Message	Date
DottaandPaperclip	1511c92965	fix(evals): account for superseded delegation wakes (#18319) ... ## Thinking Path > - Paperclip lets people ask agents to delegate work and return the result. > - The chat feature eval checks task creation, worker execution, saved output, and delivery to the source chat. > - A task can first belong to the CEO and then move to Morgan. > - Paperclip correctly cancels the old assignment wake before it starts. > - The eval treats that queue record as a failed provider run and stops before Morgan works. > - This change accounts for that record separately and preserves the complete outcome checks. > - The report also keeps partial check results when an attempt fails. ## Linked Issues or Issue Description **What happened?** The delegation journey stopped on an unstarted queue record with `issue_assignee_changed`. The CEO had created a task, assigned Morgan, saved the brief, and returned a task link. The test stopped before worker execution. Its report showed no individual checks even though the saved observation contained partial results. [Original public measurement](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/chat-progress-39e-112-latest-121-measurements-with-screenshots-40649033672/?view=case&case=chat-features.delegate-announcement). **Expected behavior** Keep the obsolete queue record and its events. Continue observing the intended worker. Pass only after three successful native work runs, a current Morgan-authored announcement, a completed task, automatic source-chat delivery, and browser readback. **Steps to reproduce** 1. Ask the CEO to have Morgan write and save a launch announcement, then return it to chat. 2. Create the task under the CEO, then assign Morgan before the queued CEO run starts. 3. Observe the cancelled old assignment. The previous eval aborts at that point. **Paperclip version or commit** Based on `6da112ab40d7b3cc0bcb79e08ddbb9b915a268a4` from `origin/master`. **Deployment mode** Disposable Product E2E fixtures. The live check uses the trusted default-branch workflow and AWS fleet. Related contribution: #17794 introduced these two journeys. The open-PR search found no separate fix for this queue-accounting defect. ## What Changed - Recognize only explicit dispatch-gate reassignment cancellations with a null start time and no native issue. Verify the company, prior CEO owner, and target task now assigned to Morgan before excluding that record from work-run counts. - Keep every queue record in the full observation and final ledger. Retain its events with an explicit reason for the absent provider log. - Let the feature flow own its complete saved-state and browser oracle. Return only actual work runs to the common provider checks. - Record saved-state checks on failures as well as successes. Capture the required `final-state.png` after the feature-specific browser readback. - Add two explicit local OpenCode cells alongside the two existing Codex subscription cells. Use the current production-qualified model. The earlier linked campaign used a different model, so this is a new measurement. - Advance the grader and suite versions. Keep the original request, output requirements, 120-second run cap, five-minute observation bound, budgets, and single-attempt policy. ## Verification - Red/green regressions cover queue accounting, missing-log collection, partial checks on evidence failure, and the required final screenshot. Flow tests invoke the actual polling callbacks: restoring the original callback reproduces both early-stop errors. Started cancellations still fail. - All 87 focused flow/grading/queue tests pass. Catalog/report checks and the 137 additional Node checks pass. Eval typechecking passes on final commit `0a6e11408c4470683417e6bf7965c73cec2a9a60`. - All current CI tests/builds pass. Greptile is 5/5 on the final commit, with no unresolved review threads. - The final commit adds only flow test coverage. Its registered definition digest equals live target `e4e0be875ee12c134fe8dd2ce15a88bfcbb100ef`: `ac6bba21d4b9cb86088be5d00df6442af90268186d6e94c9bea7d611d0488563`. - Replaying the original saved observation still fails for missing Morgan execution and delivery. Its stale queue record no longer counts as provider work. | Operation | Selected coverage | Observed result | | --- | --- | --- | | [41071710105](https://github.com/paperclipai/paperclip/actions/runs/41071710105), `b8dff816` | Backlog and delegation | Refused at the exact private-retention gate before fixtures or providers. Preserved without feature coverage. | | [41073708150](https://github.com/paperclipai/paperclip/actions/runs/41073708150), `39be9e45` | Backlog and delegation | Both passed their full product/browser oracles. Retained grades correctly fail missing `final-state.png`; the feature flow had saved the image under another name. Fixed in `e4e0be87`. Each original archive remains private and intact; later workflow concurrency superseded only the report aggregation. | | [41076854978](https://github.com/paperclipai/paperclip/actions/runs/41076854978), `e4e0be87` | Backlog and delegation | Backlog passed in 92.576 seconds, with a 47.388-second native run, complete required artifacts and passed cleanup. Delegation was interrupted by a runner-host shutdown at 15:19:23 UTC; all later evidence-upload steps were skipped. Its private archive is absent, so behavior, usage and cost remain unverified. | | [41080065077](https://github.com/paperclipai/paperclip/actions/runs/41080065077), `0a6e1140` | Delegation only | Passed on the final commit in 240.786 seconds. All 14 product checks plus browser readback passed; exact private identity, current worker output, Done state, correlated source-chat reply, required screenshot and cleanup verified. The three native runs took 42–60 seconds each. Passing control was not repeated. One-cell workflow and report enforcement passed. | The completed delegation in `41073708150` passed all 14 product checks plus browser readback in 222.593 seconds. Morgan saved the current two-sentence announcement, the task reached Done, and the CEO automatically delivered that saved text to the original chat. Its three successful native runs took 49.117, 40.060 and 39.107 seconds. The failed required screenshot remains a failed result in that measurement. All live operations use explicit IDs, concurrency one, no automatic retries, the unchanged 120-second per-run cap, and scoped private evidence retention bound to the exact workflow, target, original run, selector and launcher digest. These narrow measurements do not qualify the full catalog or establish repeated-run reliability. The broad local support suite passed 5,139 tests, with two skipped. Two database tests hit exhausted host semaphore capacity after the installed Postgres package's missing library symlinks were repaired. Their recheck reported `could not create semaphores: No space left on device`. No host IPC resources or kernel settings were changed. The corresponding full CI tests passed on a fresh host. ## Risks - A broad cancellation exception could hide failed work. This exception requires the dispatch gate's exact receipt, a null start time, the expected company and prior owner, and the actual reassigned task. Started and unproved cancellations still fail. Backlog creation permits no task wake. - New manual OpenCode cells can spend provider money only when explicitly selected. Default scheduled selection is unchanged. - This changes eval interpretation and evidence collection. It does not change product scheduling or task ownership. ## Model Used OpenAI Codex, GPT-6 family, with code execution and tool use. The exact model ID and context-window size are not exposed to 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 - [ ] 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>	2026-10-30 10:40:00 -05:00
DottaandPaperclip	2ee36a4c0a	test: add chat feature scenarios and real task workflow evals (#17794) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agent Chat lets a person assign work through a conversation. > - Feature and API maps do not prove that a natural request produces the right saved result. > - Product E2E already owns isolated fixtures, browser runs, evidence and cleanup. > - This pull request adds a chat scenario catalog and two real task workflows in that framework. > - Independent checks verify the task, its author, the saved output and the reply in chat. > - The benefit is a small executable foundation for broader chat coverage. ## Linked Issues or Issue Description **What existing behavior does this improve?** It adds real task creation and delegation coverage for Agent Chat. **Current behavior** Existing maps describe features and agent access. Existing chat evals cover specific protocols and handoffs. They do not connect a broad human-request catalog to independent saved-state checks. **Proposed behavior** Keep 244 planned scenarios across 35 feature areas. Promote two requests into the explicit-only `chat-features` suite. One saves a backlog task without starting work. The other delegates an announcement and returns the saved result to the original chat. **Reason and benefit** A successful tool call or confident reply does not prove the requested result exists. These evals check public API state, attribution and browser access after a real model run. **Breaking changes** No API or data schema changes. Chat instructions now require actual relative task links. The local E2E private-state scanner recognizes Codex account ownership metadata while continuing to reject credentials and model-visible identity leaks. **Additional context** Related prior work: #13969 and #13762. This improves verification of the existing Agent Chat roadmap item. It does not add a new core product feature. ## What Changed - Add YAML scenarios with capability IDs, starting context, user requests and expected outcomes. Validate all area files against their schema, global ID uniqueness and exact feature-map capability coverage. - Add two local native Codex subscription cells. Send the exact catalog request once through Chromium. - Seed only prerequisites through public APIs. Use shipped agent instructions and explicit disposable budget policies for unpriced subscription usage. - Check task ownership, creation audit, no premature execution, current worker-authored output, automatic completion delivery and usable links after reload. - Keep private Codex account ownership metadata distinct from bearer credentials. Scan rollout message content, all tokens and published evidence as before. - Teach normal chat instructions to use relative task links instead of invented hosts. Update tests, maps and the runbook. - Repair existing E2E contract drift: include the active Copilot identity in the image digest and update Pi profile expectations. Keep global catalog cardinality in its owning catalog test. Product E2E owns the new flow and grader. Its existing launch and report paths remain in use. The shared scanner change is limited to owner-private temporary Codex runtime metadata. Other callers retain the full secret scan. No UI, endpoint, schema or permission behavior changes. ## Verification - `node scripts/check-maps.mjs` passed: 35 areas and 191 capability IDs. - Product E2E typecheck and focused tests passed before rebase: 225 Vitest tests and 137 Node tests. Conversation service tests passed 40/40. Five existing runner turn-binding regressions passed. - Historical live measurement used real Chromium, server, database and native Codex with no automatic retries. The initial same-definition batch passed 5/6 cells. Backlog passed 3/3; delegation passed 2/3. One CEO startup failed because the runner did not receive a provider turn identity. The cause is unresolved. - An unchanged follow-up passed delegation three times and backlog twice. Combined evidence is backlog 5/5 and delegation 5/6. Each attempt passed cleanup. Later passes do not erase the earlier failure or prove a runner repair. - Earlier failed revisions exposed an invented task hostname, private account metadata false positives and missing funding for unpriced subscription usage. Those led to the specific changes above; the failed records remain retained locally. - The runbook records campaign identifiers and definition provenance. Generated result artifacts are not committed. Historical live results predate this branch rebase; fresh PR-source live evidence and final validation are recorded below. - Full workspace typecheck and production build passed on the rebased branch. Product E2E typecheck passed. - The full Product E2E support suite first reproduced four baseline failures: two missing active Copilot image-identity inputs and two stale Pi/catalog expectations. The narrow contract fixes above resolved them. The full rerun passed 4,282 Vitest tests plus 137 Node tests, with two skipped. Image checks verify mutation changes the hash, unchanged input stays stable, and an input closure without Copilot is unaffected. Original failed logs remain retained. - The repo-wide local Vitest run has reported two failures in unchanged service files. The Workspace connection test passed an isolated recheck and its full file passed 469/469 unchanged. The recovery test hit a 15-second timeout; its full file passed 89/89 unchanged. The run later hit explicit PostgreSQL `semget` resource exhaustion and was interrupted after those observed failures; no final aggregate counts or full local pass are claimed. Its complete available log is retained. No assertion or timeout was weakened. Matching remote server CI passed on the same source. - On clean PR head `39ad80b523ce194c90aafafd077d48dd9a36b4c5`, backlog passed in `local-2026-10-26T19-24-13-014Z`. The parallel delegation fixture failed during PostgreSQL initdb, before browser or model execution; its failed cleanup grade is retained. A disposable standalone initdb probe then passed. No semaphore sets or kernel settings were changed before that probe/recheck. After confirming no remaining fixture process, its private temporary files were removed. - One separate delegation recheck passed in `local-2026-10-26T19-29-24-800Z`, with one attempt, zero automatic retries and successful cleanup. Both passing current-source cells use definition `3f2c9ba2d03e3dcddde47ddab3b43bc52de977f25e256fceae0a24c9128e204b`. The earlier bootstrap failure and the historical provider startup failure remain distinct. These local results omit automatic source-SHA metadata; the clean checkout SHA is recorded here from the execution session. - The later broad local run reported `could not create semaphores: No space left on device`. A bounded cleanup removed 32 current-user PostgreSQL sets with 17 members, marker 537, a dead recorded creator and zero waiters. No live process or kernel limit was changed by that cleanup. The test process group was subsequently stopped; no broad rerun is planned. This direct later resource failure does not prove the precise cause of the earlier live initdb failure. - Review follow-up adds only static catalog validation and documentation of the existing five-minute journey bound. The executable definition is unchanged. Focused catalog/oracle tests passed 28/28; no redundant live run is claimed. - Final head `30fcb36ac9ae1f0270a589dbfafcec91ca6a03f3` passed all required and underlying CI checks in [run 40391402347](https://github.com/paperclipai/paperclip/actions/runs/40391402347), including `ci / verify` and `ci / e2e`. Greptile and Paperclip Review both scored this head 5/5, with all review threads resolved. The static follow-up leaves every executable definition input byte-identical to the live-tested source at `39ad80b`. Normal squash auto-merge is enabled. Actual merge is pending eligible code-owner approval for the owned dependency manifest; no protection is bypassed. ## Risks - Live model behavior can vary. One missing turn-start event remains unexplained; this PR does not claim provider reliability or broad catalog qualification. - Subscription costs are unpriced. Explicit fixture funding admits that usage. Run counts and deadlines remain bounded, but the dollar limit cannot cap unknown cost. - Account identity exceptions apply only to known private Codex runtime files. Token reuse, unknown files and model-visible rollout content remain blocking findings. - Deterministic checks establish saved state, output structure, required brief facts and access. They do not score all prose quality. Initial samples also require human inspection. ## Model Used OpenAI Codex, GPT-6 family, with reasoning, code execution, browser and repository tools. The precise model revision and context window are not exposed to this session. Live eval agents used `gpt-5.6-sol`; that is the evaluated model, not a claim about the authoring model. ## 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 local tests; focused suites pass and the interrupted full-run failures are documented above - [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>	2026-10-26 16:19:39 -05:00