4666 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
DottaandPaperclip	d1f3e7bb0d	docs: refresh README capabilities and roadmap (#14741) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The README introduces the product and directs people to setup and the roadmap. > - The product now supports more harnesses, connections, skills, and team workflows than the README shows. > - Some copy still describes available features as future work or makes claims broader than the implementation. > - This pull request updates the README and matching roadmap entries from current source evidence. > - Readers can see what they can use, what requires setup, and what remains experimental. ## Linked Issues or Issue Description **Issue type** Outdated information and missing documentation. **Where is the issue?** README.md feature descriptions, four pillars, setup, FAQ, and roadmap; related entries in ROADMAP.md. **What's wrong?** The README omits supported adapters and major connection and skills workflows. It presents Connected Apps and Agent Chat as wholly future work. Some budget, audit, and approval descriptions also need more precise wording. **Suggested fix** Keep the existing structure, four-pillars picture, and completed roadmap milestones. Extend the four-pillars table without removing its existing content. Add concise descriptions of supported features. Correct capability and setup claims. Distinguish available, experimental, and planned work, and explain that the roadmap follows the default branch. Searched open README and roadmap PRs and related documentation issues. Refs #14640, which proposes a separate launch-video update; this change preserves the existing video. ## What Changed - Expand adapter coverage and describe model choice alongside durable team context. - Add six feature cards: connections, personal identities for shared agents, Skill Studio, routines, artifacts and feedback, and team templates. - Describe experimental agent conversations and external chat/email entry points. - Correct budget, audit, approval, goal, session, secret, portability, and multi-organization claims. - Qualify export portability: plain environment values and local paths can remain, so packages need review before sharing. - Retain the four-pillars image and all existing table content; add connection identities, skill history, team templates, and run history. - Correct mention wake behavior, persistent npx data, source-build prerequisites, and hosting guidance. - Retain all 18 completed README roadmap milestones and its original closing sentence. Add seven completed milestones: Connected Apps, personal/shared AI accounts, Shared Agents Use Personal GitHub Identities, skill version history, document comments and revisions, company-wide search, and mixed-model/harness teams. - Keep the README’s yellow roadmap entries to their names. Expand the new milestones in ROADMAP.md alongside the existing status updates for agent chat, memory, recovery, evaluations, and queue scope. - Preserve all existing top-level section headings and their order. No runtime files change. Research: reviewed 393 candidate change summaries from a 60-day history of 1,274 non-merge commits, then checked relevant implementation, feature defaults, contracts, and current public adapter documentation. The main gaps span adapters, connections, responsible identities, skills, routines, deliverables, teams, chat/memory status, governance claims, and setup. Suggested follow-up work: refresh the product demo; add three concrete use cases; shorten Quickstart by moving secondary setup paths into the docs. ## Verification - Passed: `git diff --check`. - Passed: local link and asset targets, Markdown anchors, and HTML table nesting in both edited files. - Passed on the earlier revision: rendered README inspection on GitHub, including the new feature cards and roadmap status labels. - Passed: exact restoration checks for the original picture and both requested passages, preservation of every original pillar-table cell, and all 18 original checked roadmap items. - Passed on this revision: seven new green milestones synchronized between README.md and ROADMAP.md, three short yellow labels, local link targets, table markup, and verification that README edits are confined to its roadmap. - Verified new milestone claims against AI Connections and managed GitHub identity contracts, Skill Studio restore behavior, document annotation/revision routes, company search, and the adapter registry. - Passed on the earlier revision: `pnpm -r typecheck`. - Passed on the earlier revision: `pnpm build`. No runtime code changed in this documentation follow-up. - `pnpm test:run` reported a failure in unchanged native-session recovery code. Stopped the broader local run after reproducing it with `pnpm exec vitest run --project @paperclipai/server server/src/services/native-runtime/native-session-resume.test.ts --no-file-parallelism --maxWorkers=1`: 40 passed, 1 failed. The failing case is “archives a damaged prior epoch before completing guarded same-task replacement with real runnerd”; `retainedNativeCleanupJournalMatches` returned false at line 1125. The full local suite did not complete. - Previous revision `4bef0af3da13b03875efacd9e1f165bc69932b25` received Greptile 5/5; the export wording finding remains addressed. A fresh review is pending for this documentation follow-up. - CI remains pending. This PR remains a draft; local runtime tests are not green. - Checked feature claims against the adapter registry, feature catalog, Connections contracts and action UI, skill and team services, task semantics, deployment docs, and source startup code. - Browser suites were not run. This change edits documentation only. ## Risks Low runtime risk: only README.md and ROADMAP.md change. The default branch can lead published packages, so the roadmap now links to release notes. Provider setup and feature flags still affect availability. Live provider integrations were not requalified for this documentation pass. ## Model Used OpenAI GPT-6 via Codex. The exact model variant and context-window size were not exposed in this session. Used repository research, reasoning, shell tools, and documentation editing. No subagents were used. ## 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 - [ ] 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>	2026-09-30 13:02:58 -05:00
scotttongandClaude Opus 5.5	6aa908ef15	fix(heartbeat): cancel queued runs whose claim is permanently rejected (#14738) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The heartbeat scheduler claims queued runs for each agent in `startNextQueuedRunForAgent`, and startup recovery calls it through `resumeQueuedRuns` > - `claimQueuedRun` can throw a 4xx `HttpError` that the run's own rows decide, for example the 403 from #14043 ("Queued-message interrupt authority is unavailable") > - The claim loop does not catch that error. The run stays `queued`, so every recovery pass and every restart claims it again and gets the same error > - During periodic recovery the error is logged, but it stops the claim loop, so all queued runs of every agent after it wait. During startup recovery the error is rethrown, so the server cannot boot > - We saw this on a self-hosted instance: one card-answer run from #14043 put the server in a crash loop for about 11 hours (216 failed boots) until the run was cancelled by hand in the database > - This pull request cancels a queued run when its claim fails with a 403 that cannot clear, keeps a run queued when its 4xx rejection can clear later, and in both cases continues with the rest of the queue > - The benefit is that one bad queued run can no longer stop the server from starting or stall other agents. #14510 fixes the specific identity mismatch; this change is the safety net for this whole class of error ## Linked Issues or Issue Description Refs #14043 Refs #14510, #13539, #13315 ## What Changed - `server/src/services/heartbeat.ts`: The claim loop in `startNextQueuedRunForAgent` catches errors from `claimQueuedRun` for each run. - A 403 is a permanent rejection. In the claim path it only comes from the run's persisted identity (an unverifiable interrupt receipt, a manual wake without a user), and those rows do not change. The loop cancels that run with error code `queued_run_claim_rejected`. `cancelRunInternal` also cancels the wakeup request and releases the issue execution lock. - The cancel runs after the agent start lock is released (`.finally()`), because `cancelRunInternal` promotes the next queued run under the same lock. A cancel inside the lock waits 30 seconds for the stale-lock timeout. - Any other 4xx (for example a 422 `responsible_user_unresolved`, or a 409) can clear later. The run stays queued, the loop logs a warning, and it continues with the next run. A later recovery pass tries the run again. - Non-HTTP errors and 5xx errors (for example database errors) keep propagating, as before. Startup still fails when the database or another dependency is broken. - If a cancel itself fails, the loop logs the error. The run stays queued for the next recovery pass. - `server/src/__tests__/heartbeat-queued-run-claim-isolation.test.ts`: New embedded-Postgres tests with a mocked adapter: - A queued-comment interrupt run with an unverifiable receipt is cancelled, `resumeQueuedRuns()` resolves, and a second pass stays clean. - A claimable run behind a rejected run of the same agent is claimed and executed. - A run with an unresolvable responsible user (422) stays queued, and the claimable run behind it is executed. - A rejected run for one agent does not stop the queue of another agent. ## Verification - `cd server && npx vitest run src/__tests__/heartbeat-queued-run-claim-isolation.test.ts` passes. - Without the change in `heartbeat.ts`, all four tests fail (the 403 `Queued-message interrupt authority is unavailable` and the 422 `responsible_user_unresolved` escape `resumeQueuedRuns()`). - All `heartbeat*`, `*queued*` and `run-identity` server suites pass locally (1252/1253). The one local failure, `heartbeat-process-recovery` "redacts opaque environment-bound credentials from Sentry diagnostics", fails the same way on unchanged `master` on my machine and passes in CI. - `pnpm -r typecheck`, `pnpm test:run` and `pnpm build` pass locally. ## Risks - Behavior shift: a queued run whose claim fails with a 403 is now cancelled, not left queued. Such a run could never start before this change, so no runnable work is lost. The cancel reason and the `queued_run_claim_rejected` error code make the cancel visible on the run. - A run with another 4xx rejection stays queued, as before, but it no longer stops the rest of the queue. It logs a warning on each recovery pass until the cause clears. - Low risk otherwise: no schema, API or UI change. > 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 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 - [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>	2026-09-30 10:41:17 -07:00
DottaandPaperclip	94e8dec56b	fix(runner): preserve tool outcomes through shutdown and restart (#14734) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The runner sends authorized tool calls to the server and saves their results. > - A provider turn can stop while a server write is still running. > - The old shutdown path invented a failed result that could conflict with the real result. > - Truncated execution input and incomplete recovery records made the failure harder to diagnose. > - This pull request preserves exact inputs and actual outcomes through shutdown and restart. > - Tests force the race and crash boundaries so safe retries do not repeat writes. ## Linked Issues or Issue Description **What happened?** Stopping a turn during a server tool call could record a false failure, then reject the actual result as a conflict. The diagnostic input formatter could truncate instruction content before execution. A crash during saved-result delivery could leave that delivery permanently indeterminate. Cleanup could hide the first failure, and a retry could overwrite earlier run logs. **Expected behavior** Keep dispatched tools pending until their actual result is known. Preserve accepted input bytes. Accept identical result delivery without failing the task. Reject conflicting results with enough evidence to diagnose them. Recover saved-result delivery without repeating the business operation. **Steps to reproduce** 1. Hold an instruction update at the filesystem commit barrier. 2. Stop its provider turn before the server returns the result. 3. Release the write, deliver its result, and replay the same result. 4. Repeat with a restart before and after the delivery receipt is saved. 5. Check that there is one write and one audit row, and that the exact result survives. **Paperclip version or commit** The change was developed from `44736c9c7` and rebased onto `0e5830887`. **Deployment mode** Self-hosted server with the native runner. Tests use local runner processes, scripted providers, and PostgreSQL. Related work: #12353 added durable semantic tool receipts; #12384 added durable Codex tool recovery; #12404 bound semantic tools to ACPX sessions. #14633 covers separate native-provider cancellation and qualification work. This PR addresses server semantic-tool outcomes and their durable delivery. No duplicate fix was found. AgentMail discovery is outside this PR. ## What Changed - Close turn admission without inventing results for dispatched tools. Keep pending calls and accept late actual results. - Accept identical result replay with a diagnostic warning. Include call identity and both result hashes in real conflict errors. - Preserve exact execution arguments. Reject prohibited or oversized input before dispatch. Keep diagnostic previews redacted and bounded. - Commit instruction-attempt evidence before the filesystem write. Save completed mutation receipts so concurrent and restarted duplicates return the first result. Recheck authorization before replay. An attempt without a completed result stays unknown and cannot execute again. Definite pre-write failures save and replay their original error without another write. - Recover an interrupted saved-result delivery only for backends with durable result receipts. Never replay an ordinary business operation with an unknown outcome. - Preserve the initiating error when cleanup also fails. Record incomplete settlement evidence. Propagate typed unknown-outcome errors through the native tool wrapper without creating a false completed tool result. - Append run-log attempts and restore the durable log before appending after local file loss. Reject incomplete restores. Publish a restored prefix only if the destination is absent so concurrent attempts cannot overwrite new lines. - Add deterministic race, crash, replay, authorization, exact-content, and log-restoration tests. Document their assertions in `packages/paperclip-runner/docs/durable-recovery.md`. ## Verification - Current head: `7e088f4c7fba8ebabf98ae95485a5753b013d489`. All 55 applicable checks pass; four conditional/manual checks are skipped. This includes build, typecheck, Rust, both runner TypeScript shards, server and workspace tests, all eight browser shards, isolated runner compilation, and the clean-install release dry run. [CI run](https://github.com/paperclipai/paperclip/actions/runs/36746101110). - Greptile reviewed this exact head at 5/5 with zero new findings. All three earlier review threads are resolved. - Focused local verification includes 11 instruction integration tests, 23 surrounding authority/tool tests, 26 run-log tests, and 169 controller/driver tests. The post-rebase controller/transport/runtime selection passed 415 tests. The full Rust release suite passed 617 tests with two ignored. The real-process SIGKILL recovery test passed three consecutive runs. - The fault matrix in `packages/paperclip-runner/docs/durable-recovery.md` uses explicit barriers, real PostgreSQL rollback, durable journal reloads, and killed runner processes. It covers late results, identical and conflicting replay, exact long content, concurrent log restoration, lost commit acknowledgements, and definite failure replay after the original CAS base becomes valid again. No paid model calls are needed. - Full local recursive typecheck and build passed during implementation. Server typecheck and the runner TypeScript build passed after the review fixes. The broad local repository test run was stopped after repeated database startup timeouts. Four timing/launch failures in an earlier broad runner run passed focused reruns without changed assertions or timeouts. These are local verification limitations; the complete current-head CI suite is green. An earlier CI workspace job received an infrastructure shutdown signal; its current-head replacement passed. ## Risks - A stopped turn can remain blocked when a dispatched operation has no proven result. The system does not guess its outcome or rerun its effect. - Conflicting results still fail settlement. Existing failed or conflicting journals are not repaired automatically. - Accepted semantic input is limited to 480 KiB of encoded JSON to fit the encrypted transport. Larger input fails before execution. - Instruction filesystem writes and database receipts are not one atomic storage operation. A separately committed attempt and audit record survive rollback. An attempt without a completed success or definite pre-write failure receipt remains blocked as an unknown outcome. It is not replayed or reported as success. - Run-log restoration now reads the durable object before appending when the local log is missing. Failed or incomplete reads reject the append. - No schema migration, dependency change, workflow change, or AgentMail change is included. ## Model Used OpenAI Codex, GPT-6, with reasoning, tool use, code execution, and test analysis. The exact served model ID 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 (focused suites; the broad local run limitation is recorded 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>	2026-09-30 11:54:56 -05:00
Nicky LeachandPaperclip	af5c2d101c	fix(paperclip-runner): deliver the shutdown settlement event past the terminal-turn gate (#14668) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The OpenCode driver maps provider events to runtime request events > - A provider turn can end before a pending runtime request receives its answer > - The consumer reads one turn's events and stops at that turn's terminal event > - A settlement event that arrives after that terminal event never reaches the consumer > - This pull request settles the request inside its own turn, before the terminal event > - The benefit is reliable request settlement without weakening late-frame protection ## Linked Issues or Issue Description **What happened?** A pending runtime request stayed open after an OpenCode turn failed through `session.error`. Session shutdown then dropped its settlement event as a late provider frame. **Expected behavior** The driver must deliver `runtime_request.expired` with the original `turnId` and `itemId`, inside the same single read pass the consumer performs on that turn. **Steps to reproduce** 1. Start an OpenCode turn that creates a native runtime request. 2. Leave the request pending and fail the turn through `session.error`. 3. Close the session and inspect the emitted events. **Paperclip version or commit** `22d41c6081f05658e0d7c8485d0f22f35af4a79e` **Deployment mode** Built from source with the OpenCode driver test fixture. ## What Changed - Settle a pending runtime request as soon as its own turn goes terminal, before the terminal turn event. - Add an optional `bypassTerminalTurnGate` parameter to the OpenCode session emitter, and set it on the settlement emit. - Keep a settlement loop in session close as a fallback for a request whose turn never went terminal. - Add a fixture trigger and a regression test that reads one turn in a single pass. - Keep the late provider frame gate unchanged for every provider event path. ## Verification - Run `pnpm vitest run packages/paperclip-runner/src/drivers/opencode/opencode-server-driver.test.ts`. - Run `tsc -p tsconfig.json --noEmit` in `packages/paperclip-runner`. - Run `tsc -p tsconfig.surfaces.json --noEmit` in `packages/paperclip-runner`. - Confirm that the regression test receives `runtime_request.expired` with the original identifiers. - Confirm that the late-frame tests still report dropped provider frames. ## Risks Four call sites now reach the settlement path: session close and the three terminal turn paths (completed, cancelled, and failed). At the three terminal turn paths the turn is still the active turn, so the gate admits the settlement event with or without the parameter. Session close is the only place where the parameter changes the result of the gate, and only for a request whose turn already ended. Every provider event path keeps the existing terminal-turn gate. The known driver test failure is pre-existing and does not touch this change. ## Model Used Claude Sonnet 5, with code execution and test assistance. ## 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 the changed tests locally; the known pre-existing failure remains documented above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation, or no documentation change applies - [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 have addressed all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-09-30 09:48:14 -07:00
DottaandPaperclip	3c561642b4	fix(chat): resolve approvals and preserve unanswered questions (#14613) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents ask for decisions and optional details through cards in chat. > - A clear approval in a message can leave the matching card pending. > - An unanswered question can also block an unrelated later reply. > - Decisions need a saved source message, while optional questions need to remain answerable in history. > - This pull request records conversational decisions and lets users move on from questions and answer them later. ## Linked Issues or Issue Description **What happened?** Native Claude and Codex could act on approval in chat while the original approval card stayed pending. Pending question forms stayed above the composer, were absent from history, and could suppress later chat replies. A late native question answer could wait for a finished run to reconnect. **Expected behavior** The active agent records a clear approval or refusal against the exact card and user message. Ambiguous replies do not grant consent. Users can send another message without answering a question. The question remains pending in history and can be reopened and answered later. The saved answer reaches the agent. **Steps to reproduce** 1. Ask an agent to propose work with a confirmation card, then approve it in chat. 2. Check that the original card records that approval before work starts. 3. Ask an interactive question, send an unrelated message, and reload. 4. Open the unanswered question from history and submit an answer. Related work: #14408 added completion delivery. #14607 tests completion reporting turns. Neither records conversational answers on approval cards. ## What Changed - Add a confirmation endpoint backed by a user comment, with schema validation, OpenAPI discovery, and native Plan-mode access. Ask mode remains read-only. - Check company, active run, actor, current session, message provenance, revision, and resolver policy. Save the decision and audit in one transaction. Retries do not repeat effects. Emit resolution telemetry after commit. - Give fresh and resumed chat turns the actual pending confirmation identities. Teach agents to save clear conversational decisions before acting and to clarify ambiguity. - Keep unanswered Agent Chat questions as compact history entries. A newer user message closes the old form. Question cards never contribute to composer pending counts or navigation, including after dismissing a fresh form. The history card is the sole reminder; clicking it restores that exact form and draft. - Preserve Agent Chat questions when later messages or questions arrive. Historical ordinary inputs no longer gate later chat replies. Current-run requests, task execution, and governed approvals keep their gates. Remove the special acknowledgement-publication proof helpers that this rule replaces. - Route answers to finished native runs through durable fresh-wake delivery, with existing idempotency and source-question context. Settle late replies against contiguous completed conversation turns and freeze their history replay; failed, unhandled, and newly arriving messages remain actionable. - Add real-component Storybook scenarios, database and UI regressions, and a three-turn native Claude/Codex E2E case. Capture distinct, UI-ready screenshots and report the individual assertions. ## Verification - Focused decision/publication/UI regressions after merging master: 288 passed; subsequent UI draft, failed-send, and conversation checks: 199 passed. - Native question and durable delivery regressions: 106 passed, including all four terminal run states and exactly-once late delivery. Seven targeted regressions fail against the original implementation and pass with the fix. - Latest conversation/decision/native-delivery regressions after the master merge: 121 passed. Covers completed progress, missing or failed intervening turns, new messages during a late reply, stale sessions, and frozen retry/replay boundaries. Four new assertions fail before the ordering fix. - E2E support suite after the master merge: 792 passed. Negative controls reject expired cards, wrong questions/answers, stale or missing replies, unrelated clarification forms, and unexpected tasks. - The embedded-browser walkthrough caught one additional defect: dismissing a fresh question still showed a composer badge. Both Cancel and close-button regressions failed before the fix. The fix at `65f2ade12` passes 170 chat-thread tests and 792 E2E support tests. After merging master, 232 chat-thread/confirmation tests, server/UI typechecks, and token gates pass. The preview and two-provider live E2E pass at `e5512a206`; Greptile is 5/5 with zero unresolved threads at that commit. All 55 checks are now successful at `e5512a206` (four conditional checks skipped), including the aggregate verification gate and clean-install canary test. The first attempt was interrupted by simultaneous CI worker shutdowns; one failed-job rerun passed without code changes. - [Published Storybook](https://d1p6rlowie26tp.cloudfront.net/storybook/branches/codex~2Fchat-approval-resolution/?path=/story/chat-comments-agent-chat-unanswered-questions--moved-on): nine real-component scenarios. Manually exercised move on, reopen, preserve draft, answer later, answer one of multiple questions, and a custom mobile answer in the embedded browser. Retested fresh Cancel and close-button dismissal in the updated build, then reopened and submitted the preserved Green selection and inspected its answered receipt. Static preview has no live model/backend; its callbacks are fixture responses. - [First live campaign](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36714504406-1/) reproduced the late-answer completion-state defect on both providers despite correct saved answers and acknowledgements. It also exposed a valid imperative clarification rejected by the old oracle. Both issues are fixed with regression controls; this failing run is retained as evidence. - [Four-cell qualification](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36717804064-1/) passed 4/4 at `2bf8a1009`: unanswered-question return and ambiguous confirmation, each on native Claude and Codex. Inspected saved state, source-message decisions, visible cards, and agent replies. Both late-answer chats settled to waiting; no unrequested tasks were created. [Final branch rerun](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36719666238-1/) passed 2/2 at `142630720`: the same unanswered-question journey after merging master, plus an additional screenshot and browser assertion for the actual late-answer acknowledgement. - [Composer-reminder E2E](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36727006818-1/) passed 2/2 at `5b62c52d9`: native Claude and Codex, three turns each, with explicit no-badge assertions before and after reload. Inspected saved pending/answered state, both screenshots with a clear composer, and actual Blue acknowledgements; all five behavioral matchers passed per provider and neither created tasks. Cost coverage is partial; this is bounded workflow qualification. - [Fresh-dismissal E2E](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36742773318-1/) passed 2/2 at `e5512a206`: native Claude and Codex, including fresh Cancel, clear composer, reopen, unrelated message, reload, late Blue answer, and actual agent acknowledgement. All five behavioral matchers pass per provider. Inspected the fresh-dismissal screenshots and saved pending/answered identity; neither created tasks. Cost coverage is partial (4/6 runs). - Prior evidence remains available in [the earlier campaign](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-36642252725-1/). Its early loading screenshot and overwritten final capture prompted the UI-ready, distinct screenshot fixes. ## Risks - The model interprets intent. The server verifies permission and provenance; it does not infer consent from text. Ambiguous and unrelated replies are not approvals. - Historical questions can accumulate. They remain visible, pending, and answerable; no automatic answer or expiry is invented. - The change to completion gates is scoped to Agent Chat and ordinary historical inputs. Current-turn and governed approvals retain their existing controls. - Live qualification is limited to the selected stories. Broader native onboarding finalization remains separate work. - No database migration. Telemetry adds no fields or values; the contract and README document the commit boundary. Privacy review was requested on the PR. ## Model Used OpenAI Codex, GPT-6 family, with reasoning, repository tools, code execution, and browser-test orchestration. 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 - [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>	2026-09-30 11:46:46 -05:00
DottaandPaperclip	0e5830887b	perf: reveal task content sooner and parallelize issue reads (#14727) ... ## Thinking Path > - Paperclip helps people manage AI agents and their work. > - A task page must show saved replies quickly so a person can read the work. > - The title could appear while the conversation waited for unrelated metadata and transcripts. > - Thread requests also waited for enriched task details, while the server read several independent fields in sequence. > - This change shows saved content as soon as it is ready, starts thread reads earlier, and runs independent server reads together. > - Native event history takes priority over legacy log fallback, and mentioned tasks load on intent. > - Content-free timing spans make the remaining server delays visible without recording task content. ## Linked Issues or Issue Description **What happened?** Task titles and properties appeared quickly, but saved conversation content stayed hidden for several more seconds while metadata and run transcripts loaded. **Expected behavior** Saved replies and the task description should be readable without waiting for supporting history. Returning to a cached task should show content within a frame or two. **Steps to reproduce** 1. Open a task with saved comments and completed runs. 2. Delay the task activity and runs responses by five seconds in the browser. 3. Observe whether saved content remains hidden until those responses finish. 4. Navigate away and return to the task to check cached navigation. **Paperclip version or commit** The change was developed from `1b48e73e0` and rebased onto `44736c9c7`. **Deployment mode** Built from source, tested in an authenticated staging deployment and with local response replay. Related: #14667 overlaps the transcript reveal behavior and adds separate retry UX. This PR also changes navigation prefetch, parent metadata gates, native log fallback, server read scheduling, and timing spans. #12647 proposes a separate SQL predicate optimization in the runs service. #13597 and #13095 are earlier loading fixes. ## What Changed - Start activity and runs when the task page mounts, alongside task details and comments, using the route reference for shared query keys. Hover/focus prefetch does not start full history reads. - Reveal saved comments and descriptions while metadata and transcripts load. Preserve strict waits for linked-comment navigation and tasks with only runtime content. - For settled native runs, fetch legacy logs only when event history is empty or fails. Preserve live-log subscriptions for queued and running native runs. Fetch mentioned-task details on hover or focus. - Run independent issue-detail enrichment and run metadata reads in parallel while preserving recovery dependencies. - Add `Server-Timing` phases and opt-in OpenTelemetry spans, plus regression tests and observability documentation. ## Verification - **313 tests passed** across the initial seven focused component, cache, timing, and scroll suites. After review fixes, **313 tests passed** across five task-page, cache, prefetch, and live-transcript suites (`pnpm exec vitest run` with `--maxWorkers=1`; these sets overlap). - `pnpm -r typecheck` and `pnpm build` passed locally before the final UI-only review fixes. UI typecheck/build and `pnpm check:token-gates` passed after those fixes. The final commit also passes full typecheck and build in CI. - The full local Vitest run was attempted. Several unrelated embedded PostgreSQL fixtures failed to start, and parallel test workers hit timeouts. Focused reruns passed. A later local full-suite rerun was stopped after the complete CI suite passed; it is not claimed as a local full-suite pass. - Final commit `d1e147145`: **54 checks passed, 2 skipped**, including all server/workspace test shards, all eight browser E2E shards, full build/typecheck, Runner checks, release registry, and canary clean-install verification. [CI run](https://github.com/paperclipai/paperclip/actions/runs/36734642034). Greptile **5/5** after two reviews; both findings fixed and all review threads resolved. No merge conflicts. - Live browser tests on an existing task with two saved replies: median full reload to visible content fell from **1.92 s** (3 samples) to **1.40 s** (5 samples). Cached return fell from **421 ms** (1 sample) to **29 ms** (3 samples). These are observed samples, not a performance guarantee. - With activity and runs delayed by five seconds, saved content appeared in **1.38 s** on desktop and **1.33 s** on mobile. The inspected comment did not move when metadata arrived. Verified history expansion, task properties, pending-input navigation, dashboard return, and mobile layout. - A separate local replay with fixed responses reduced visible-content time from **5.12 s** to **2.15 s**. This isolates frontend behavior and is not a live-server benchmark. ## Risks Progressive history can change the thread after first paint. Existing anchor behavior is retained and covered by tests and delayed-response browser checks. Query aliases must stay aligned for invalidation. Parallel reads can increase short bursts of database work; dependent recovery operations remain ordered. Full reloads still depend on network and task-detail latency. No schema or authorization change. OpenTelemetry remains disabled without an operator endpoint. The added spans use a closed set of phase names and carry no task IDs, task content, or exception text. I checked `ROADMAP.md`; this is a performance fix within the existing task page. ## Model Used OpenAI GPT-6 in Codex. The runtime does not expose a more specific model ID or context-window size. The agent used reasoning, code editing, terminal tools, and Chrome performance profiling. ## 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>	2026-09-30 10:26:28 -05:00
DottaandPaperclip	a36cbffa9e	fix(connections): broaden natural-language and aggregator search (#14725) ... ## Thinking Path > - Paperclip manages agents and the services they need for work. > - Agents use connection search to discover a setup path before they request access. > - Tool-only filtering hid channel and AI methods from this search. > - Requiring every query word to match rejected normal task descriptions. > - A small aggregator index also omitted supported apps such as Circleback. > - This pull request broadens retrieval and returns purpose-specific setup guidance. > - Agents can choose a relevant result while existing access and provider-choice checks still apply. ## Linked Issues or Issue Description **What happened?** A search such as “AgentMail create an email address and manage an agent mailbox” returned no usable result. “Help me find tools for circle back” also missed Composio's supported Circleback toolkit. Queries longer than 200 characters failed validation. **Expected behavior** Return useful native and verified aggregator matches from natural-language queries. Include channel/email methods when Chat connectors is enabled. Identify each method's purpose and the correct setup path. **Steps to reproduce** Use the queries above with `connections_search` from an active task. Enable Chat connectors for the AgentMail case. The regression suite reproduces these misses before the change. Related routing work: #13941. This change does not change the runner failure path or add channel setup to tool-only connection cards. ## What Changed - Rank name and capability matches. Accept extra words, split names, small spelling errors, and queries up to 4,000 characters. - Include tool, channel/email, and AI methods. Return company-prefix setup links for channel and AI flows. - Add a dated snapshot of 1,583 official Composio toolkit names and a refresh script. Merge duplicate MCP variants for search and link each support claim to official evidence. - Find authorized indexed aggregator namespaces within longer queries. Return multiple app matches when the agent needs to choose. - Prefer exact app names over fuzzy matches for other apps; retain existing AI readiness. - Preserve native preference, company and identity boundaries, administrative denials, and saved provider consent. - Add relevance and database regressions, extend native tool-authority coverage, and document search behavior. ## Verification - Red: 15 new assertions failed against the previous implementation; the existing baseline passed. Added red-green regressions for Motion versus fuzzy Notion and existing AI access during review. A further regression covers mixed ready/unconfigured AI results and their per-result setup guidance. - Green: all 103 tests in the eight focused shared, database, runtime-tool, fixture, and route suites pass on the latest commit. - `pnpm -r typecheck` and `pnpm build` passed. - Latest-commit CI passed: 54 successful checks and two skipped checks, including the full test matrix, browser E2E, typecheck, build, and canary dry run. - The long local `pnpm test:run` attempt began before the review fixes and retained transformed pre-fix search code; it also hit an unrelated timing failure. Fresh serial reruns of the affected search suites and three timeout cases passed all 124 tests. Parallel local route shards hit two additional database setup timeouts; both suites passed all 17 tests on a fresh serial rerun. The complete corresponding CI suites also passed. Duplicate broad local runs were stopped after CI completed. The local UI suite independently passed all 7,007 tests. - Greptile: 5/5 on `cc6a0180d`; all review findings resolved. - Browser verification passed in a disposable local instance through real process-agent search requests: AgentMail opened its setup flow with the requester selected; the saved Circleback choice produced the Composio setup card; a paragraph-length Notion query produced its setup card. No provider credentials or external accounts were created. - The browser test caught an invalid UUID-based setup URL. The fix uses the company prefix and has a regression assertion. - A 3,971-character catalog query found Circleback first in a local 10 ms spot check after sharing query preparation across the catalog scan. This is a single measurement, not a performance guarantee. ## Risks - Broader retrieval can return extra candidates. Named services rank first; agents must select the relevant method. - The public support snapshot can age. It proves catalog support, not account authorization or the availability of every requested action. - Channel and AI methods use existing setup links. The tool connection card still accepts tool methods only. - No schema, migration, credential, or runner lifecycle changes. ## Model Used OpenAI Codex (GPT-6). The session does not expose a more specific model identifier or context-window size. Used reasoning, repository search, code execution, tests, and browser tools. ## 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>	2026-09-30 10:20:50 -05:00
DottaandPaperclip	dd7fc1f90a	fix: raise the native journal read limit to 256 MiB (#14711) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Native sessions persist control-plane state so they can resume safely. > - The state includes committed provider history needed for recovery. > - The server, runnerd recovery, and durable control plane validate this state before trusting its identity. > - Their differing 64 MiB and 192 MiB limits can reject a valid journal before recovery. > - This pull request aligns all three local state limits at 256 MiB. > - Larger files remain bounded, while recovery can read larger valid histories. ## Linked Issues or Issue Description Refs #13882 Refs #14312 ## What Changed - Raise the server and runnerd recovery limits from 64 MiB to 256 MiB, and align the durable control-plane limit from 192 MiB to 256 MiB. - Add coverage for a valid history above 64 MiB and rejection above 256 MiB. ## Verification - Matching server recovery passed with more than 64 MiB of actual committed event payloads (128 events with 512 KiB deltas). - The real runnerd exact-authority resume regression with the test Codex provider passed with 193 MiB of valid JSON whitespace appended. It crosses the former 192 MiB core limit and confirms the same provider identity. This exercises the runner process and durable control plane with a simulated provider, not a live OpenAI API call. This test used approximately 1.15 GiB peak RSS. - The actual runnerd reader accepted valid 256 MiB JSON and rejected valid 256 MiB + 1 byte. The reader call took 231 ms; the fresh process peaked at 1,244 MiB RSS. - Server tests reject mismatched identity above 64 MiB and files above 256 MiB. - `pnpm -r typecheck`, `pnpm build`, and `git diff --check` passed. - Full local `pnpm test:run`: 13,730 passed, 575 skipped, 7 failed across 6 files. All failures were embedded PostgreSQL startup errors after five attempts. They affected agent hiring, instruction revisions, environment images, reviewed chat bindings, issue monitoring, and legacy continuation authority. The focused journal tests passed; the latest pushed head passed all ordinary CI checks. Superagent is the only blocking check. ## Risks - **Open review concern:** Greptile is 5/5, but Superagent is `ACTION_REQUIRED` with two P2 findings on the server and runnerd readers. Both flag the increased synchronous parsing and memory cost. This PR keeps the requested fixed-limit change small. It does not add a worker parser or a process-wide memory budget. This resource tradeoff needs review before merge. - Large state parsing is synchronous and can consume several times the file size in memory. - Remote checkpoint archive and expanded-size limits remain 64 MiB, so this change alone does not make larger remote checkpoint transfers portable. ## Model Used - OpenAI Codex, GPT-6, with delegated assistance from `gpt-6-luna` at high reasoning effort; tool use and code execution. The GPT-6 context window is not exposed in this task runtime. ## 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 - [ ] 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>	2026-09-30 10:08:25 -05:00
DottaandPaperclip	44736c9c7c	fix(ui): align runner commentary with chat replies (#14716) ... ## Thinking Path > - Paperclip helps people manage AI agents and their work. > - Agent chat shows run commentary and the agent's reply in one thread. > - The runner commentary text starts 4px left of the following reply. > - The runner activity group omits the gutter that the reply bubble uses. > - This pull request gives runner commentary the same token-based gutter. > - Both text blocks now share one left edge. ## Linked Issues or Issue Description **What happened?** In agent chat, the first commentary line of a completed run starts about 4px left of the following agent reply. This is visible in the Codie conversation. **Expected behavior** Runner commentary and ordinary agent reply text should have the same left edge. **Steps to reproduce** 1. Open an agent chat with a completed run that has commentary and an agent reply. 2. Compare the first commentary line with the following reply at the left edge. 3. Inspect the wrappers. Before this change, `task-chat-phase-interstitial` has zero left padding and `task-chat-agent-bubble` has 4px. ## What Changed - Add the existing `px-1` spacing token to runner commentary. - Add a regression test that compares the runner commentary gutter with the ordinary agent reply gutter. ## Verification - `pnpm check:token-gates` - `pnpm --filter @paperclipai/ui exec vitest run src/components/task-chat/TaskChatRunnerActivityGroup.test.tsx src/components/task-chat/TaskChatTurn.test.tsx src/components/task-chat/TaskChatBubble.test.tsx` - `pnpm --filter @paperclipai/ui typecheck` - `pnpm --filter @paperclipai/ui build` - A full repository typecheck and build passed earlier on this branch. All latest-commit CI gates passed, including the server shard after a transient preview readiness failure was rerun. - Browser render: runner commentary and reply paragraphs both start at x=32px after the change. ## Risks Low risk. This changes only the horizontal padding of runner commentary. Long commentary lines may wrap 8px sooner. The full local test suite was stopped after the code changed during its run; the focused UI tests and latest-commit CI suite passed. > I checked `ROADMAP.md`. This bug fix does not duplicate planned core work. ## Model Used OpenAI GPT-6 in Codex. The runtime did not expose a more specific model ID or context window size. The agent used reasoning, shell tools, and read-only browser inspection to diagnose the layout and verify the 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 --------- Co-authored-by: Paperclip <noreply@paperclip.ing> canary/v2026.930.0-canary.6	2026-09-30 09:48:41 -05:00
DottaandPaperclip	d30b03bd8c	test: add persistent E2E coverage for human blocker decisions (#14707) ... ## Thinking Path > - Paperclip manages work for AI agents. > - Agents use the coordination skill when work needs human authority or a scope decision. > - PR #14188 replaced automatic manager escalation with direct blocker handling. > - This behavior needs real browser, server, database, and provider tests. > - The test must verify saved human input, task ownership, and resumed work. > - This pull request adds six reusable Product E2E cases and improves the skill examples that they exercise. ## Linked Issues or Issue Description Refs #14188. The merged change needs repeatable behavior coverage. The new suite tests missing administrator access, missing hiring permission, and requester scope questions. Searches found no duplicate blocker-guidance suite. This extends the existing eval system described in ROADMAP.md. ## What Changed - Add the explicit-only `blocker-guidance` Product E2E suite. It has three local scenarios on legacy Codex and legacy Claude. - Use the production UI and public APIs to create work, save a human-only question or confirmation, answer it after reload, and resume the same task. - Check requester identity, ownership history, manager activity, hiring, saved answers, and completion. Keep direct text input as a separate UX result. - Save pending and final screenshots, API checkpoints, skill hashes, provider evidence, and billing data through the existing report pipeline. - Isolate the Claude fixture home. Verify the served skill bytes before dispatch so an old installed skill cannot silently replace the evaluated skill. - Improve the coordination and hiring skill examples. Include the human-only policy, requester address, wake behavior, and handling of authorized scope changes. - Grader v5 requires the exact approved public welcome note as a new worker comment. Browser input checks reject unwritable scope cards before clicking, and confirmation direction must be saved in the resolution before the worker wakes. - Add grader calibration and browser-input tests. Update the fixture guide and generated capability inventories. ## Verification - `pnpm build`: passed after rebasing onto current master. - `pnpm -r typecheck`: passed. - `pnpm test:e2e:runner:typecheck`: passed. - `pnpm test:e2e:runner:unit`: 742 tests passed. - `pnpm test:e2e:runner:browser-support blocker-input.spec.ts`: 10 tests passed. - `pnpm test:e2e:runner -- --list --suite blocker-guidance`: six cells found. - Capability inventory and generated-contract checks: passed. - `pnpm exec vitest run server/src/__tests__/hiring-operational-examples.test.ts`: four tests passed after synchronizing the generated API reference and section anchor. - Full general and serialized test suites: passed in CI on `6652cee74517039676bad6a720f213625d265acd`. The redundant local `pnpm test:run` was interrupted after complete CI coverage passed; it is not claimed as a completed local full-suite run. - Final GitHub checks: 54 passed, two optional Storybook checks skipped. The runtime-exposure startup test hit a 10-second readiness timeout once, passed a targeted local reproduction, and its CI shard passed the single retry without code changes. - Current-head Greptile: 5/5, clean check, zero unresolved threads. - Historical live measurement on September 29 at `4edc77ae2b95b10dd61426ce3f042bac00527ad9`: three independent six-cell runs scored 5/6, 6/6, and 6/6. Claude Sonnet 4.6 passed 9/9. Codex `gpt-5.6-sol` passed 8/9. These runs predate this rebase. - Version 5 changes the scope answer to an exact approved publication draft. The historical runs do not qualify that new requirement; the two-provider scope pilot at `49a1f4eab369948b9e3b34a6ce436489e875e4ec` passed Codex and failed Claude. Claude posted the correct salary-free sentence but omitted its required reference line from that comment, placing the reference in a separate completion message. The `public-welcome-note` check correctly failed. An earlier Claude database-startup failure was retained separately; its fresh-instance retry reached the model. This pilot is not a six-cell qualification. - The failed Codex scope case omitted `addresseeUserId`. The strict routing check remains. All 18 attempts had clean evidence manifests and passed cleanup. - To repeat with provider credentials: `pnpm test:e2e:runner -- --suite blocker-guidance --max-parallel 1`. This is a paid, opt-in suite and is excluded from `--all`. ## Risks - The live suite measures variable model behavior. The retained 17/18 historical result and the current 1/2 scope pilot are not all-pass qualifications. These paid cases are opt-in; their observed model failures remain visible independently of deterministic CI checks. - A separate generic task-replacement diagnostic still exposed a Claude refusal. The ordinary cases use specific business decisions. The diagnostic is not a standalone catalog case in this change. - Earlier measurements included an old installed Claude skill and test defects. Their grades remain retained and are not combined with the three final repetitions. - Skill examples can affect when agents ask for human input. Downstream permission checks still apply. - Native runners, Daytona, agent-requester routing, and real external connection authorization are outside this suite. - Raw provider traces and credentials remain private. No screenshots, raw reports, secrets, workflow changes, or lockfile changes are committed. ## Model Used OpenAI GPT-6 through Codex assisted with this change. The exact deployed variant and context window size are not exposed in this session. The assistant used reasoning, repository edits, tool use, and shell execution. The evaluated models were `gpt-5.6-sol` and `claude-sonnet-4-6`. ## 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>	2026-09-30 07:56:54 -05:00
DottaandPaperclip	1b48e73e0b	feat(ui): add secondary navigation for agent chat (#14706) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agent Chat already provides a persistent conversation with each agent. > - Its shortcuts share the primary navigation and do not give chats a dedicated place. > - People need to find agents, start a chat, and switch conversations without moving the page layout. > - This pull request adds a secondary chat sidebar and a landing page around the existing chat surface. > - The same conversation, composer, history, and context panel remain in use. ## Linked Issues or Issue Description Refs #13283 and #13420. This extends the existing experimental Agent Chat navigation after review of the component and page stories. It supports the CEO Chat roadmap item through the existing task-backed conversation model. **Subsystem affected** The board UI and the company-scoped conversation list API. **Current behavior** Chat shortcuts sit inside the primary navigation. There is no dedicated landing page with a searchable conversation list. A separate landing header also moves the sidebar when an agent is selected. **Proposed behavior** Show a Chat entry in primary navigation. Keep a searchable agent sidebar beside the chat content. The plus button starts or reopens the current user's single conversation with that agent. Keep the header and sidebar in the same positions before and after selection. **Reason and benefit** People can find agents and return to persistent conversations without leaving the chat area or creating duplicate chats. **Breaking changes** The experimental chat navigation changes. Explicitly adding a chat now resolves its conversation immediately. Direct visits to unused agent chat URLs remain read-only. The existing per-agent routes and message contracts remain compatible. No database migration is required. ## What Changed - Add an account- and company-scoped conversation list endpoint with the existing access checks, feature gate, and OpenAPI entry. - Add the live secondary sidebar, landing page, avatars, search, loading states, errors, and retry controls. - Make the agent picker wait for chat creation and display failures. Existing agents reopen the same conversation. A dismissed selection cannot close a reopened picker or navigate over a newer choice. - Preserve recent-activity ordering and terminated agents’ chat history. Scope live list refreshes to the current user’s conversation events. A failed historical-agent lookup leaves healthy chats usable and offers a focused retry. - Keep the sidebar and header stable across chat routes. Keep mobile selection in the navigation drawer. - Use the production components in Storybook. Prepare the theme and mobile viewport before mounting the page to avoid the startup flash. - Update product documentation, the design guide, and navigation tests. Replace old browser expectations for stars and recent shortcuts with persistent conversation and layout coverage. ## Verification - `pnpm -r typecheck`, `pnpm build`, and `pnpm check:token-gates` pass after rebasing onto master. - Focused UI tests pass, including 105 sidebar, picker, and live-update checks after review fixes. The 33 conversation service and route tests and 10 OpenAPI checks pass, including ownership, feature gating, and concurrent creation. - The full local test run passed 14,163 server tests before three environment or timeout failures. The embedded Postgres startup, connector socket, and native runner failures all passed direct reruns. - Browser test-drive verification covers a real provider reply, add and reopen, persisted history after reload, no-match search recovery, mobile drawer dismissal, and top-aligned context panels. - Browser measurements confirm that the sidebar has the same position and dimensions on the landing page and an agent conversation. - Storybook builds and its add-and-reopen interaction passes. - The revised browser regression passes locally against a freshly built throwaway instance. It covers stable sidebar geometry, add/reopen uniqueness, drafts, search, history, and terminated-agent history after reload. The full CI browser suite also passes. - Latest commit `b323577d9523180104df4000eaceedea2772608c`: all 54 completed checks pass, including the complete server/workspace/browser suites, aggregate verification, build/typecheck, security scans, and canary packaging. The two Storybook jobs are skipped by their workflow conditions. [CI run](https://github.com/paperclipai/paperclip/actions/runs/36714052050). - Greptile reviewed this same commit at 5/5 with no remaining actionable findings; all review threads are resolved. - Reviewer path: enable Agent Chat, click Chat, use plus to choose an agent, send a message, switch away, and reopen that agent. One conversation must remain, with its history intact. ## Risks - The new sidebar lists persistent conversations instead of starred and recent shortcuts. - Chat creation is asynchronous. Errors stay visible in the picker, and delayed responses cannot navigate into a previous company or account. - The shell adjustment is limited to chat routes and preserves the existing conversation implementation. ## Model Used OpenAI GPT-6 through Codex, with reasoning, code execution, and browser tools. The session does not expose the exact API model ID or context-window size. ## 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 #123` / `Refs #123` 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>	2026-09-30 07:35:32 -05:00
DottaandPaperclip	d72389bee2	feat: add Browser Use Cloud connector and live task browsers (#14627) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The Apps gateway gives agents governed access to external tools. > - Browser Use Cloud can run browser work, but a tool result alone does not let a person watch or take over. > - A task needs a durable browser session, a visible viewer, and recorded costs. > - This pull request adds a Browser Use Cloud v4 connection and interactive browser tabs on tasks. > - People can follow the work, interact with the page, and retain the browser after the agent finishes. ## Linked Issues or Issue Description **Problem or motivation** Agents need governed access to Browser Use Cloud. People need to see and interact with the same browser from the task. A browser must remain available after a run finishes and appear at the correct point in the task feed. **Proposed solution** Add a native REST connection for the v4 API. Bind each session to its company, task, agent, and credential grant. Open its interactive viewer in the task side panel. Record provider costs as financial events. Use `browser-use-cloud` as the app and connector key. Keep its skill with the connector and deliver it only with authorized connection tools. **Alternatives considered** A v3 MCP connection would expose tools without the v4 lifecycle integration. An external viewer link would leave the task. A fixed viewer size would prevent pages from responding to changes in the task pane. **Roadmap alignment** This extends the governed Apps gateway and Connected Apps roadmap. It uses the existing task, grant, secret, approval, and financial records. The work was requested by the maintainer. A search found no duplicate Browser Use connector PR or issue. ## What Changed - Add the Browser Use Cloud app, brand asset, API-key connection, and profile settings under the `browser-use-cloud` key. - Bundle the `browser-use-cloud` skill with the connector. Keep it out of global `skills/` discovery. Deliver it only with authorized task/run connection tools. Remove retired connector skill keys from runtime overlays and preserve unrelated browser skills. - Expose seven v4 tools through the governed gateway and deliver them to native and CLI agents. - Persist sessions, browsers, runs, event and recovery cursors, shutdown leases, and cumulative cost accounting. Recover uncertain paid starts without replaying them. - Enforce task ownership, credential grants, approvals, revoked access, and budget limits. - Add interactive task browser tabs and compact chronological feed entries. Retain the viewer across tab switches and keep visible idle browsers open. - Add debounced automatic viewport fitting, standard size presets, and a viewer ownership lease. - Add lifecycle, authorization, accounting, viewport, UI, and Storybook coverage. - Add an idempotent database migration after the current master migration. Preserve deployed migration hashes. Migrate pre-release Cloud connection and financial keys without replacing grants, credentials, or browser history. - Document provider behavior, live acceptance results, and the lack of documented passkey forwarding. ## Verification - Full workspace typecheck and production build pass on the updated branch. - Token gates, brand asset validation, module boundaries, and migration ordering pass. - Cloud tests verify global skill exclusion, authorized task/run delivery, unassigned agents, disabled connections, revocation, adapter isolation, and secret exclusion. The existing AgentMail connector assignment test also passes. - Migration replay runs twice against existing browser work and financial records. It preserves the records and avoids duplicate costs. - The focused provider, app catalog, OpenAPI, connection gateway, and migration regression suites pass. Recovery coverage includes lost replies, process crashes, provider rejection, and browser arrival acknowledgement. - All 54 checks pass on `2974b5f03641ad0cea3c941d8c02579316fa8c92`, including the full test matrix, browser E2E shards, build, typecheck, security, and release canary. Two optional Storybook jobs are skipped. - Greptile is 5/5 on the same commit, with zero unresolved review threads. The corrected review uses the actual master-to-head diff. - The local `pnpm test:run` started and was stopped after the full CI matrix passed. It did not complete locally; the full-suite result above comes from CI. - Earlier live acceptance used an isolated company with a capped provider credential. The agent opened paperclip.ing, the embedded viewer accepted navigation, and the same browser stayed available after completion and tab switches. - The local Storybook build passes. Stories cover the panel, footer, feed entries, settings, lifecycle failures, and viewport modes with an offline viewer fixture. ## Risks - Browser Use charges for hosted work. Provider caps and local budget checks reduce exposure; reported costs can arrive after work completes. - Viewer and CDP URLs grant access to the browser. The server validates and restricts them. They are excluded from agent results and durable event data. - Runtime resizing of v4 agent browsers uses a provider option confirmed by live testing but absent from its published agent schema. Resizing during a click may invalidate coordinates. Fixed presets remain available. - Viewport ownership is process-local and resets on restart. The lifecycle and accounting records remain in the database. - The original intermittent embedded-viewer stall has not been fully diagnosed. A bounded reconnect and active-session recovery cover the observed failure paths. - Live tests did not cover every revocation, approval, rate-limit, or restart case. Deterministic integration tests cover those paths. Passkey forwarding is not claimed. - Unknown create outcomes keep the credential available for cleanup. Run-list absence cannot prove a paid POST was rejected, so recovery stays pending until it can identify provider work. ## Model Used OpenAI Codex, GPT-6. Used reasoning, repository search, code execution, browser interaction, and test tools. The exact serving model ID and context-window size were 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 - [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> canary/v2026.930.0-canary.5	2026-09-30 07:13:16 -05:00
DottaandPaperclip	f4f9a7c613	test(runner): guard continuation after journals exceed 2 MiB (#14312) ... Add an actual runner resume regression above the former 2 MiB journal boundary and an explicit-only three-turn Daytona workflow that grows real execution history. Verify journal size and distinct completed tool calls without exporting private payloads. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-30 07:01:45 -05:00
DottaandPaperclip	2f6fa3b6dc	fix: recover provider authentication inside tasks (#14629) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents need a working model provider connection to run a task. > - A provider can reject a stored credential after the task starts. > - The failed run must ask the responsible user to repair that connection. > - This pull request adds that request directly to the task and reuses Connections sign-in. > - The user can choose an API key or subscription, then continue the task with a fresh session. ## Linked Issues or Issue Description **What happened?** A run that ended with `acpx_auth_required` or another known provider authentication error did not immediately offer an inline way to connect the provider. A repair form could also lock the user to the failed account's sign-in method. **Expected behavior** Show a provider connection card in the task as soon as the authentication failure is saved. Allow the responsible user to connect or repair the provider with any supported sign-in method. Keep the connection name automatic and resume the task after successful setup. **Steps to reproduce** 1. Run a task with a supported provider and an expired or invalid credential. 2. Let the run fail with a provider authentication error. 3. Open the task and attempt to repair the connection. Related work: Refs #13724 and #13726. This change adds the inline task repair flow and method choice. ## What Changed - Classify provider authentication failures and create one connection request for the current task. A persisted blocked classification suppresses automatic retries only after the repair card is created; unsupported providers retain their existing recovery path. - Mark only the attributed, unchanged credential as needing sign-in. Preserve credentials that were refreshed after the failed run started. - Reuse the provider sign-in controls inside the task. Allow API key and subscription choices for Claude, Codex, and Grok. Keep names hidden and generate a default from the user, provider, and method. - Keep the existing account when reconnecting with the same method. Create and select another account when the method changes. Validate updates to explicit agent bindings through the normal agent save path. - Require explicit adoption for legacy agent authentication. Validate in the agent environment, then commit the binding, connection install, audit, and card completion in one transaction. Keep failed setup and account selection visible and retryable. - Add regression tests and update the specification and Connections documentation. ## Verification - Fresh local verification: 199 tests passed across the inline form, provider method selector, default naming, authentication and recovery classifiers, run liveness, OpenAPI routes, database adoption/rollback, and Cursor execution suites. The adoption database suite also passed against disposable Docker PostgreSQL. - Full repository `pnpm build` and `pnpm -r typecheck` passed on the latest commit. Token gates are clean. - Embedded browser: opened real task cards from seeded authentication failures; switched Claude from API key to subscription and back; switched Codex from subscription to API key; confirmed the name field stays hidden. Provider sign-in was not completed with real credentials. - The broad local `pnpm test:run` started before review fixes and was interrupted after the working tree changed; it is not counted as a passing full run. Fresh focused tests passed. CI supplies the full test and browser suite results for the current commit. - CI is green on commit `4b97a4e447045ff3d7516525a187a5d1d21e0d4c`: 54 checks passed and two Storybook checks were skipped by their path rules. The workspace preview job passed on one rerun after a local-server startup timeout; its rerun passed 835 tests. - Greptile is 5/5 on the same commit with no actionable findings and no unresolved review threads. ## Risks - Incorrect authentication classification could prompt for a connection unnecessarily. Tests exclude tool authorization, quota, and unrelated runtime failures. - A method change selects the new personal provider default, which also applies to other agents that use that user's default. Explicit account bindings use the existing permission and runtime validation path. - Credential invalidation must not race with refresh or reconnect. The code compares the saved credential generation and grant update time under locks. - No database migration or new credential storage format is required. ## Model Used OpenAI GPT-6 through Codex. The exact model ID and context window size were not exposed in this session. Capabilities used: reasoning, repository editing, shell commands, database tests, and embedded-browser interaction. ## 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 (focused tests listed 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> canary/v2026.930.0-canary.4	2026-09-30 06:39:37 -05:00
3 people	a027f76a72	fix(test): wait for the completion reporting turn before asserting chat status (#14644) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - Agent chat tests verify the state changes that follow delegated work > - A delegated task wakes its source conversation with a completion reporting turn > - That turn can move the conversation out of `in_review` for a short time > - The test read status during that turn and failed under runner load > - This pull request waits for the completion turn to reply and settle before it reads status > - The benefit is a stable test that checks the intended final state ## Linked Issues or Issue Description **What happened?** The end-to-end test read a conversation status immediately after a delegated task finished. The completion reporting turn briefly changed the status during its checkout. The test could read `in_progress` instead of the final `in_review` state. **Expected behavior** The test should read the conversation status after the completion reporting turn replies and the conversation settles. **Steps to reproduce** 1. Run `tests/e2e/agent-chat-projects.spec.ts` on a loaded four-vCPU runner. 2. Run the test named `shared questions resume and existing project reuse creates no project card`. 3. Observe that the status read can race the completion reporting turn. **Paperclip version or commit** `master` at the commit under test. **Deployment mode** Local dev with the end-to-end test runner. **Installation method** Built from source. **Agent adapter(s) involved** Not adapter-specific. **Database mode** Not database-related. **Additional context** The deterministic reproduction failed 3 of 3 times before this change and passed 3 of 3 times after this change. ## What Changed - Wait for the completion reporting turn with the existing `idle` helper before the status assertion. - Add a comment that explains the race and the required settling point. - Change only `tests/e2e/agent-chat-projects.spec.ts`. ## Verification - Run `tests/e2e/agent-chat-projects.spec.ts` with Playwright. - Confirm that all 16 tests pass. - Run the deterministic reproduction before and after the change. - Confirm that the target test passes. - Confirm that the end-to-end shard and the full CI suite pass. - Confirm that Greptile gives a 5/5 result with no open findings. ## Risks Low risk. The change affects one end-to-end test. It adds one wait through an existing helper. It does not change product code, raise a timeout, add a fixed sleep, or skip a test. ## Model Used OpenAI Codex, GPT-5, tool use and code execution. The model context window and reasoning mode were not provided for this handoff. ## 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> Co-authored-by: Claude Sonnet 5 <noreply@anthropic.com> canary/v2026.930.0-canary.3 nightly/v2026.930.0-nightly.0	2026-09-29 22:09:37 -07:00
Nicky LeachandPaperclip	eb31b926a1	fix(runner): keep the OpenCode session event stream open across turns (#14582) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - Paperclip Runner keeps provider sessions and their event streams for agent runs > - The OpenCode driver closed its event queue after each terminal turn > - A second turn on the same session then lost its response and completion events > - This pull request keeps the queue open between turns and rejects late events for sealed turns > - The benefit is reliable multi-turn OpenCode sessions with visible diagnostics for late provider events ## Linked Issues or Issue Description **What happened?** The OpenCode driver closed its event queue when a turn completed, was cancelled, or failed. A second turn on the same session then lost its response and completion events. **Expected behavior** The session must keep its event stream open between turns. Each turn must deliver its response and one terminal event. The session must close the stream only during session shutdown or an unrecoverable pump error. **Steps to reproduce** 1. Start one OpenCode session. 2. Run one turn and wait for its terminal event. 3. Run a second turn on the same session. 4. Confirm that the second turn delivers its response and terminal event. **Paperclip version or commit** `c0e1d87ddc181329471fa80a2061b2c538bb6618` **Deployment mode** Built from source with the Paperclip Runner package test suite. ## What Changed - Keep the OpenCode event queue open across completed, cancelled, and failed turns. - Track sealed turn ids and reject later events for those turns with a diagnostic event. - Preserve queue shutdown on session close and unrecoverable pump errors. - Give each simulated fixture turn unique provider event ids. - Add regression tests for completed, cancelled, failed, and closed-session paths. ## Verification - Type check: `cd packages/paperclip-runner && node ./node_modules/typescript/bin/tsc -p tsconfig.json --noEmit` passed. - Driver tests: `cd packages/paperclip-runner && npx vitest run src/drivers/opencode/opencode-server-driver.test.ts` passed except for the known pre-existing flaky test described below. - Consumer tests: `cd packages/paperclip-runner && npx vitest run src/native-session-runtime.test.ts src/backends/harness-driver-backend.test.ts src/cli/opencode-app-server-proxy.test.ts src/conformance/harness-driver.test.ts` passed. - The known flaky test reproduced on unmodified `master` because fixture event order depends on a local MCP HTTP round-trip. - CI must run the full pull request suite. ## Risks - The queue now retains sealed turn ids for the session lifetime. OpenCode does not reuse turn ids, so this set grows with the session. - A late provider event cannot reach a later turn. The driver emits a diagnostic event so the rejection remains visible. - The change does not alter session shutdown or unrecoverable pump error handling. > For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and discuss it in `#dev` before opening the PR. This pull request fixes an OpenCode Runner bug and does not add a roadmap feature. ## Model Used OpenAI Codex, GPT-5, tool use and code execution; Anthropic Claude Sonnet 5 also assisted with the implementation. The runtime did not provide a context window size. ## 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> canary/v2026.930.0-canary.2	2026-09-29 22:05:41 -07:00
Devin Foley	f38b5693f6	fix: always enable keyboard shortcuts (#14643) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The web UI has keyboard shortcuts for the inbox, task lists, cases, and task detail, plus global shortcuts such as `c`, `/`, `?`, `[`, and `]` > - Shortcut enablement was an instance-wide General setting until #14141 moved it to a per-user preference that defaults to off > - The move did not carry the old instance value over, so every existing user lost shortcuts on upgrade and had to find a new toggle under Profile settings > - A toggle that only turns off a standard, input-safe feature costs a setting, a database column, two API routes, and a React context for little benefit > - This pull request removes both the instance setting and the personal preference and enables keyboard shortcuts for every signed-in user > - The benefit is one less thing to configure, no silent loss of shortcuts on upgrade, and less code to maintain ## Linked Issues or Issue Description Refs #14141 (the change that introduced the personal preference). **What existing behavior does this improve?** Keyboard shortcuts in the web UI stay off unless each user turns them on in Profile settings. **Subsystem affected** Web UI shortcuts, Profile settings, instance general settings, the `/api/auth/preferences` routes, and the `user` table. **Current behavior** Shortcuts default to off per user. #14141 moved the toggle from Instance settings → General to Profile settings and did not carry the old instance value over. Users who had shortcuts on lost them after the upgrade and had to find the new toggle. **Proposed behavior** Keyboard shortcuts are always enabled for every signed-in user. There is no instance setting and no personal preference. Shortcuts already ignore key presses inside text inputs and modal dialogs, so an opt-out is not needed. **Reason and benefit** Fewer settings, no silent loss of shortcuts on upgrade, and removal of a database column, two API routes, a query hook, and a React context that existed only to gate this feature. **Breaking changes** `GET` and `PATCH /api/auth/preferences` are removed. `PATCH /api/instance/settings/general` no longer accepts `keyboardShortcuts`; that schema is strict, so the key now returns 400. `instance.general.keyboardShortcuts` is no longer a valid `PAPERCLIP_HIDDEN_SETTINGS` key; the parser ignores unknown keys with a warning. ## What Changed - Removed the Keyboard shortcuts section from Profile settings, the `useUserPreferences` hook, `queryKeys.auth.preferences`, and `authApi.getPreferences` / `authApi.updatePreferences`. - Removed `GeneralSettingsContext`. The inbox, legacy inbox, task list, legacy task list, cases, and task detail pages no longer gate their key handlers. - Removed the `enabled` option from `useKeyboardShortcuts`. The app shell always registers the global shortcuts. - Removed `GET` and `PATCH /api/auth/preferences`, their OpenAPI entries, and the `currentUserPreferencesSchema` / `updateCurrentUserPreferencesSchema` validators. - Removed `keyboardShortcuts` from `InstanceGeneralSettings`, the general settings zod schema, the settings service defaults, and `HIDEABLE_GENERAL_SECTIONS`. - Added migration `0289_drop_user_keyboard_shortcuts`, which drops `user.keyboard_shortcuts`. - Updated `AGENTS.md`, `doc/SPEC.md`, `doc/SPEC-implementation.md`, and `docs/deploy/environment-variables.md`. - Parsed the stored general settings row with `instanceGeneralSettingsSchema.strip()` in the feedback vote path, so a retired key left in the row cannot reset the sharing preference to `prompt` and overwrite the stored choice. - Kept every bare global shortcut (`c`, `?`, `[`, `]`, `/`) out of open modal dialogs in `useKeyboardShortcuts`; only `/` had that guard before. - Updated the affected tests and added a Profile settings test that asserts the toggle is gone, a hook test for the modal dialog guard, and a feedback service regression test for the retired-key case. ## Verification - Typecheck passes for `@paperclipai/shared`, `@paperclipai/db` (including the migration numbering and safety checks), `@paperclipai/server`, and `ui`. - `pnpm exec vitest run server/src/__tests__/instance-settings-routes.test.ts server/src/__tests__/openapi-routes.test.ts server/src/__tests__/auth-routes.test.ts server/src/__tests__/sentry.test.ts` → 119 passed. - `pnpm exec vitest run ui/src/components/Layout.test.tsx ui/src/pages/ProfileSettings.test.tsx ui/src/pages/IssueDetail.test.tsx ui/src/pages/Inbox.test.tsx ui/src/pages/Cases.test.tsx ui/src/hooks/useKeyboardShortcuts.test.tsx ui/src/pages/Agents.test.tsx ui/src/pages/InstanceGeneralSettings.test.tsx` → 286 passed. - `pnpm exec vitest run packages/shared/src/settings-visibility.test.ts` → 16 passed. - `pnpm exec vitest run ui/src/hooks/useKeyboardShortcuts.test.tsx` → 7 passed. - `pnpm exec vitest run server/src/__tests__/feedback-service.test.ts` (embedded Postgres) → the new retired-key test passes with the fix and fails without it. - Manual: sign in with no settings changed, open the inbox, press `j` and `k` to move the selection, press `?` to open the cheatsheet. Open Settings → Profile and confirm there is no Keyboard shortcuts section. ## Risks - The migration drops a column. It uses `DROP COLUMN IF EXISTS`, and the column has no readers after this change. If you roll back to a build from before this PR after the migration has run, re-add the column first: `ALTER TABLE "user" ADD COLUMN "keyboard_shortcuts" boolean DEFAULT false NOT NULL;`. The older build's ORM selects that column when it loads users. - Any external client that still sends `keyboardShortcuts` to `PATCH /api/instance/settings/general` receives a 400. No in-repo client does. - Stored `instance_settings.general.keyboardShortcuts` values are stripped on read and ignored. - Users who never turned the toggle on now get shortcuts. The handlers skip text inputs, contenteditable regions, and modal dialogs, so typing is unaffected. ## Model Used Claude Fable 5.1 (`claude-fable-5-1`) in Claude Code, with extended thinking and tool use. ## 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 canary/v2026.930.0-canary.1	2026-09-29 21:28:06 -07:00
Devin FoleyandPaperclip	0b12ca9532	fix(server): retain context for unconfirmed adapter stops (#14639) ... ## Thinking Path > - Paperclip must keep ownership of work until termination is verified. > - A Stop request waits for the adapter and its cleanup to settle. > - After 60 seconds, an unconfirmed Stop raises an error. > - The error currently lacks run, adapter, and runtime context. > - This makes it difficult to investigate which stop path is stuck. > - This change adds bounded diagnostics while preserving termination checks. ## Linked Issues or Issue Description **What happened?** An adapter can remain unsettled after its Stop request. The resulting error says termination is unverified but does not identify the adapter or run in error monitoring. The optional Sentry setup does not capture request context, so the endpoint alone cannot fill the gap. **Expected behavior** Keep the Stop unconfirmed and preserve its live execution owner. When optional Sentry is enabled, attach enough bounded context to investigate the affected run. **Steps to reproduce** 1. Register an adapter execution control and abort its controller. 2. Leave its settlement promise pending. 3. Wait for the configured Stop timeout. 4. Observe that Stop still fails, but the event now includes the run UUID, built-in adapter, native/legacy runtime, timeout duration, and abort-requested flag. **Paperclip version or commit** Master commit `17780751551b3bc1c2521f7694026c34534c46c9`; reproduced with fake timers and mocked optional error monitoring. **Deployment mode** Server execution control, including local and hosted runs. Reporting remains opt-in. Searched related Stop PRs. #14523 and #14244 address Hermes cancellation contracts; this change only adds diagnostics to the shared unconfirmed-stop timeout. ## What Changed - Use a typed timeout error with the existing message, name, and timer stack. - Pass run/adapter/runtime identity from the cancellation owner. - Add an event-local, allowlisted Sentry context without changing the default fingerprint. - Rebuild the reported exception so arbitrary provider fields cannot be serialized. - Test timeout ownership, delayed settlement, privacy boundaries, and absence of context on unrelated events. - Document the additional opt-in fields. ## Verification - `pnpm exec vitest run server/src/services/adapter-execution-control.test.ts server/src/__tests__/sentry.test.ts`: 38 passed, five real-SDK checks skipped because the optional package is not installed. - `pnpm -r typecheck` passed; server typecheck passed again after the SDK test addition. - With audited optional `@sentry/node@10.71.0` installed only in local test dependencies, `PAPERCLIP_REQUIRE_SENTRY_TEST_SDK=1 pnpm exec vitest run server/src/__tests__/run-failure-sentry-real-sdk.test.ts server/src/services/adapter-execution-control.test.ts server/src/__tests__/sentry.test.ts`: all 45 tests passed. The real SDK uses an in-memory transport; no Sentry requests are sent. - The broad local `pnpm test:run` command did not complete in the available verification window and was stopped; no full local-suite pass is claimed. `pnpm build` passed. All sharded GitHub CI checks passed on the final PR head. - Tests use fake timers and a mocked Sentry package; no provider or monitoring requests. ## Risks This is diagnostic coverage, not a claim that the underlying stop delay is fixed. Unknown adapter/runtime values become `unknown`; malformed run identifiers become `null`. No stop reason, prompt, output, provider response, credentials, or arbitrary error properties are sent. Timeout, cancellation acknowledgement, live-owner retention, and retry behavior remain unchanged. No schema changes. ## Model Used OpenAI Codex (GPT-6), with reasoning, repository inspection, and command execution. The session does not expose a more specific model revision or context-window size. ## 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 described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run the targeted tests locally and they pass; full checks are in progress - [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> canary/v2026.930.0-canary.0	2026-09-29 16:38:02 -07:00
Devin FoleyandPaperclip	35a4448c02	fix(cursor): select failure diagnostics after trace notices (#14636) ... ## Thinking Path > - Paperclip manages work performed by AI agents. > - The Cursor CLI adapter turns process output into run results. > - Cursor can print a trace-file notice before a real error. > - The adapter used the first stderr line as the failure summary. > - This could hide the error behind an informational file path. > - This change selects the first diagnostic after that known notice and preserves the full logs. ## Linked Issues or Issue Description **What happened?** When Cursor exits with a nonzero code, a leading `cursor-retrieval: tracing to ...` notice can become the error summary. A later error remains in stderr but is absent from the summary. If the notice is the only output, the summary does not explain that the process exited unsuccessfully. **Expected behavior** Prefer the structured error, then a stderr diagnostic, then the exit code. Keep the run failed and preserve the original logs. **Steps to reproduce** 1. Use a fixture Cursor executable that writes the trace-file notice to stderr. 2. Write `Authentication failed` on the next line, then exit with code 7. 3. The old adapter reports the trace-file notice. This change reports the authentication error. 4. Repeat with only the notice. This change reports `Cursor exited with code 7`. **Paperclip version or commit** Reproduced against master commit `17780751551b3bc1c2521f7694026c34534c46c9` with local process fixtures. **Deployment mode** Local CLI adapter. The diagnostic helper is also used by environment probes. Searched open Cursor PRs and issues. PRs #14631 and #14435 concern native ACP support; #11106 concerns MCP configuration. None changes this legacy CLI diagnostic selection. ## What Changed - Skip only the exact trace-location notice when choosing a diagnostic line. - Remove terminal control codes from summary candidates. - Use the same selection for execution and environment probes. - Preserve structured-error priority, exit status, retry decisions, and raw stdout/stderr. - Add child-process regression tests and narrow parsing cases. Document the behavior. ## Verification - Two execution regression cases failed on the previous implementation; structured-error priority already passed. - `pnpm exec vitest run packages/adapters/cursor-local`: all 16 tests passed across five files. - `pnpm --filter @paperclipai/adapter-cursor-local typecheck` passed. - `pnpm -r typecheck` passed. - All GitHub CI checks passed on the PR head. One preview-runtime readiness test failed on the first attempt; its full local suite passed (28 tests, three skips) and the failed CI shard passed on retry. No unrelated source change was needed. - The broad local `pnpm test:run` command did not complete in the available verification window and was stopped; no full local-suite pass is claimed. The full sharded GitHub CI suite passed. `pnpm build` passed. - Tests use local fixture processes. They make no Cursor provider requests. ## Risks A future Cursor notice format may no longer match and will remain visible. Retrieval error lines and unknown diagnostics remain visible. This improves diagnosis; it does not claim to fix an unknown provider or machine failure. There are no schema, authentication, cancellation, or retry-policy changes. ## Model Used OpenAI Codex (GPT-6), with reasoning, repository inspection, and command execution. The session does not expose a more specific model revision or context-window size. ## 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 described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run the targeted tests locally and they pass; full checks are in progress - [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>	2026-09-29 16:36:36 -07:00
Vyacheslav ScherbininandClaude Opus 5.5	3b34d45a92	fix(ui): make destructive-foreground readable in the light theme (#14328) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The board UI uses shadcn theme tokens from `ui/src/index.css`. > - Delete and remove dialogs confirm with a red button: `bg-destructive text-destructive-foreground`. > - In the light theme, `--destructive-foreground` has the same value as `--destructive`. The label is red text on a red fill, so it is not visible. > - Operators cannot read the button that deletes data or removes a connection. > - This pull request sets the light `--destructive-foreground` to the near-white value that the dark theme already uses. > - The benefit is that every destructive confirm button shows a readable label in both themes. ## Linked Issues or Issue Description Fixes #14327 I searched open and closed PRs for `destructive-foreground`, "Remove connection" and related terms. I found no PR that changes this token. ## What Changed - `ui/src/index.css`: in `:root`, `--destructive-foreground` changes from `oklch(0.577 0.245 27.325)` (the same as `--destructive`) to `oklch(0.985 0 0)`. This is the value in `.dark`. - Added `ui/src/components/DestructiveThemeTokens.test.ts`. The test converts both tokens from OKLCH to sRGB and computes the WCAG contrast ratio. `:root` must reach 4.5:1 (AA for normal text). `.dark` must reach 3:1, because it keeps the brighter red chosen in the gallery review (about 3.6:1 today). The fix applies to all five buttons that use `text-destructive-foreground`: remove connection (`Browse.tsx`, `Connections.tsx`), delete folder (`FolderControls.tsx`), delete environment (`CompanyEnvironments.tsx`) and remove chat endpoint (`ChatEndpointDetail.tsx`). No other file uses the token. ## Verification - `pnpm exec vitest run src/components/DestructiveThemeTokens.test.ts` in `ui`: 2 tests pass. The light theme contrast is about 4.8:1. - Before the fix, the test fails for `:root` (1:1, red on red). With a light gray foreground `oklch(0.88 0 0)` it also fails (3.3:1 < 4.5:1). - Manual check on a self-hosted instance in the light theme: Apps → connection → **Remove connection** showed a red button with no visible label. After the fix, the label is white and readable. Before and after, light theme (the same classes as the confirm button: `bg-destructive text-destructive-foreground`): <img src="https://gist.githubusercontent.com/gentslava/3aa0d498e2475dd44e12bed11e7d7862/raw/c793570ebd416fe04cc390df8554515b4207590d/remove-connection-before-after.png" alt="Remove connection button before and after the fix" width="520"> ## Risks - Low risk. The change is one token in the light theme. Only the five confirm buttons above read `--destructive-foreground`. ## Model Used - Claude Opus 5.5 (`claude-opus-5-5`) in Claude Code, with tool use (shell, file edits). It found the cause, wrote the fix and the test, and wrote this description. ## 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 @cryppadotta @devinfoley a tiny one-line theme fix: the delete buttons in the light theme are red-on-red right now, so people can't see what they are about to click. Would be great to get it in when you have a minute 🙏 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com> canary/v2026.929.0-canary.9	2026-09-29 16:34:26 -07:00
DottaandPaperclip	fc36d1ccd3	test(e2e): qualify large Daytona Git workspace continuation (#14316) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A task can keep one agent process and workspace across human review turns. > - Large workspaces must survive each transfer between Daytona and the host. > - Small fixtures do not cross the previous 32 MiB filename-output limit. > - Correct files alone do not prove that finalization and recovery have settled. > - This pull request adds an explicit three-turn test with 60,000 files and independent host checks. > - The test detects lost files, replaced processes, and stale finalization retries. ## Linked Issues or Issue Description Refs: #14253, #14314, #14315, #14402, #14420. This test covers the workspace Git streaming fix in #14253 and the Daytona archive validation and native finalization fixes consolidated from #14315 and #14314 into #14402. These runtime changes are merged into master. This PR adds regression coverage. ## What Changed - Add one explicit-only native Codex Daytona test. Broad matrix runs do not select it. - Create 60,000 small untracked files through ordinary provider execution. Independently check all host file contents and the 39,828,890-byte filename manifest after each turn. Each continuation changes all generated file contents, so stale host copies fail. - Check spaces, newlines, leading hyphens, Unicode, and glob characters in filenames. - Require committed native finalization, successful workspace receipts, no active transfer or runless cleanup, and no scheduled recovery before each continuation and after the final turn. - Use a fixed external instruction bundle so the same runner PID and process identity can continue across the three browser-driven turns. Managed agent folders intentionally stop the process for file collection after #14420. Set the 20-minute idle window on the environment, then verify the admitted policy on every run. Set a 25-minute Daytona auto-stop window for this large-file test. - Save public workspace-operation evidence when an E2E attempt fails. - Bound this large-file fixture to 15 minutes per turn and 50 minutes total, reserving five minutes outside the turns for setup, host verification, and cleanup. A measured CI continuation succeeded in 11 minutes 5 seconds, exceeding the ordinary warm fixture's 10-minute deadline. The ordinary fixture and all file, process, finalization, and cleanup assertions remain unchanged. ## Verification - Final PR head: `b7eed5517426507f5d7912edb8ae49163d1783b0`. - `pnpm test:e2e:runner:unit` — 56 files and 708 tests passed locally. The catalog retains all 407 existing cells and adds this one explicit-only cell. - `pnpm test:e2e:runner:typecheck` — passed locally. - [Final-head CI](https://github.com/paperclipai/paperclip/actions/runs/36640920233) — all gates passed, including repository typechecks, tests, build, browser suites, and canary dry run. The PR has 54 successful checks and two expected skips. Fresh Greptile reviewed all seven files at 5/5; all review threads are resolved. - [Live single-cell Daytona verification](https://github.com/paperclipai/paperclip/actions/runs/36637283902) — passed on the first attempt in 1,822,299 ms (30m 22s) on `227074b7b`, using native Codex `gpt-5.6-sol` and the verified Daytona image. All three turns independently verified every one of the 60,000 file contents, 39,828,890 filename bytes, and five unusual names. All runs committed with one stable runner PID/process fingerprint, native/provider sessions, runner instance, and sandbox. Final browser/download assertions, all nine matchers, explicit cleanup, and report publication passed. Results are published through the [Product E2E history](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/) and [eval hub](https://pages.paperclip.ing/evals/). - The final `b7eed5517` follow-up only extends the total allowance from 45 to 50 minutes, updates its catalog assertion/version, and documents the setup/cleanup margin. Per-turn limits and behavioral assertions are unchanged from the successful live run; that paid run was not repeated for this allowance-only follow-up. - The [first integrated-head live attempt](https://github.com/paperclipai/paperclip/actions/runs/36632127364) is retained: turn 1 passed, then turn 2 hit the old 10-minute deadline while finalizing. Its trace records successful completion after 11m 5s and successful cleanup. Process diagnostics also showed the intentional managed agent-folder stop boundary introduced in #14420. These observations motivated the larger turn budget and fixed external instructions. - The broad local `pnpm test:run` began alongside the build and encountered server setup and port-test failures. The affected setup suites and assertions passed on focused reruns after the build, using canonical macOS temporary paths where needed. The redundant broad local run was stopped; full-suite success is established by final-head CI, not by that local run. ## Risks - The live test makes provider calls and creates a billable Daytona sandbox. It runs only when explicitly selected and deletes its sandbox during cleanup. - The test creates 60,000 files and can take several minutes per transfer. Its longer retention window applies only to this fixture. - The test will fail on a runtime that does not include all three required fixes. ## Model Used OpenAI GPT-6 in Codex assisted with code, terminal tools, and test analysis. The exact serving model suffix and context-window size were not exposed to 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> canary/v2026.929.0-canary.8	2026-09-29 17:49:36 -05:00
DottaandPaperclip	1778075155	fix(server): continue unfinished tasks after status replies (#14626) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Native runs report their task outcome through a structured finish result. > - A Board comment can permit a passive wait for the next response. > - That exception accepted reports that also admitted blocking unfinished work. > - The task then stayed In Progress without a runner, and recovery treated the wait as healthy. > - This pull request rejects that contradiction and uses the existing bounded continuation path. > - Real wait conditions and protection against obsolete requests remain in force. ## Linked Issues or Issue Description **What happened?** An agent answered a status inquiry with `yielded` and `response_wake`. The same result listed blocking remaining work. The server accepted an indefinite wait without a question, approval, dependency, or pause. No further run was queued. **Expected behavior** Unfinished ordinary tasks must continue or have a recorded reason to wait. A status reply alone must not suspend the work. **Steps to reproduce** 1. Add a Board status inquiry to an unfinished assigned task. 2. Submit a successful native result with `yielded`, `response_wake`, and `remainingWork[].blocksCompletion: true`. 3. Leave the task without any real wait condition. 4. Observe that the old policy preserves In Progress with no continuation and suppresses recovery. **Paperclip version or commit** Reproduced in the native status policy at `da887ea3e`. The branch is based on current master. **Deployment mode** Authenticated server with the native Paperclip Runner. Related work: Refs #13338 (native response waits and recovery). Refs #12071 (separate legacy recovery and retry-state work). This change fixes the native unfinished-response-wait exception. ## What Changed - Reject contradictory finish reports while the provider can still correct them. - Route accepted unfinished response waits through the existing one-follow-up continuation budget. Repeated incomplete results create a visible recovery action. - Preserve questions, approvals, dependencies, pauses, conversation lifecycles, and superseded Board requests. - Recheck the current Board source in the decision transaction before queuing repair. - Let normal recovery reconsider old committed waits that report blocking work and still have a current source. - Add policy and database regression tests. Document the rule. ## Verification - `pnpm exec vitest run server/src/services/native-runtime/status-arbiter.test.ts server/src/__tests__/heartbeat-process-recovery.test.ts --no-file-parallelism`: 346 tests passed, no skips. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - `git diff --check origin/master...HEAD`: passed. - Final head `8f0824d9bb4d631f2347f8afffe5fff6d574cb69`: all CI checks green (54 passed, 2 intentionally skipped), including all general tests, serialized server suites, runner tests, eight E2E shards, and the canary dry run. - Greptile: 5/5 on the final head, with no unresolved review threads. - The broad local `pnpm test:run` encountered an unrelated timing failure in `workspace-runtime.test.ts` (waiting for managed process-tree listeners). That test passed on an isolated rerun. The duplicate broad run was stopped; the complete CI suite passed on the final commit. - The transaction-race regression reproduced an obsolete continuation before the fix. All 12 source-change cases now pass, covering passive waits, corrective continuations, and exhausted-repair decisions. ## Risks - Agents that previously parked unfinished ordinary tasks must now continue or record an actual wait condition. - Old contradictory waits become eligible for normal recovery. Existing ownership, budget, pause, and supersession checks still apply. - The rule uses the structured blocking-work flag. It does not infer omitted work from prose. - No schema, dependency, or UI change. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository inspection, code editing, and test execution. The session does not expose a more specific deployment identifier or context-window size. ## 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>	2026-09-29 16:51:50 -05:00
DottaandPaperclip	e912f0df53	fix(ui): open text attachments in task tabs (#14297) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agent work often ends with a Markdown or plain-text file. > - Task attachments currently open outside the task panel. > - Users need to inspect those files while keeping the task conversation in view. > - This pull request opens text attachments in task tabs and adds rendered, raw, and download controls. > - The same controls work in the mobile task drawer. ## Linked Issues or Issue Description **What happened?** Opening a text attachment did not put its content in a task tab. Markdown files had no in-task rendered/raw toggle. **Expected behavior** Open Markdown and text attachments in one reusable task tab. Show Markdown as rendered content or raw text. Download the original file. **Steps to reproduce** 1. Upload a Markdown file and a plain-text file to a task comment. 2. Open each attachment from the task conversation or artifact list. 3. Switch Markdown between Rendered and Raw. Download both files. 4. Repeat at a mobile viewport width. Related work: #14193 controls artifact tab arrival. This change adds text attachment content tabs. ## What Changed - Route text attachment opens from conversation and artifact cards into task tabs. - Add a text attachment panel with accessible Rendered, Raw, and Download controls. - Preserve ordinary links for other file types. - Support the selected attachment in the mobile drawer. - Keep text-tab actions on the current rich artifact cards, including CSV previews. - Render attachment image references and diagram source without loading media URLs. - Add browser regression tests, component tests, Storybook examples, and usage documentation. ## Verification - Full workspace typecheck, production build, Storybook build, and UI token gates pass locally. - All 6,960 UI tests pass. The additional media regression passes against the real Markdown renderer and fails before the fix. CSV coverage verifies direct downloads and text tabs after preview. - Both desktop and mobile browser cases pass locally. They check rendered/raw Markdown, literal plain text, reusable tabs, review controls, and exact original download bytes. The local fixture used a separate database port because an existing socket occupied the default range. - The full CI test matrix passes on `82be5efbef26927b237a031725bb3d7fa79f637f`. The duplicate local `pnpm test:run` was stopped after this CI result; it did not complete locally. - Greptile is 5/5 on the final commit. All review threads are resolved, and the security scan passes. - All 54 final-head checks pass, including the canary dry run. The two optional Storybook jobs are skipped. ## Risks - Text attachment links now open in the task panel. Other content types keep their existing link behavior. - File display still depends on the existing authenticated attachment route. There are no API or database changes. - Raw text is displayed as text, including strings that look like HTML. Rendered Markdown keeps media references inert. ## Model Used OpenAI Codex, based on GPT-6, with code execution, browser testing, and subagent tool use. The runtime does not expose an exact serving model variant or context-window size. Recovered earlier implementation changes were reviewed and tested; their exact model metadata is unavailable. ## 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> canary/v2026.929.0-canary.7	2026-09-29 16:40:10 -05:00
DottaandPaperclip	81a52eb740	fix(ui): allow touch scrolling in new-task pickers (#14599) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The new-task composer lets users select an assignee and override its model. > - On phones, these selectors open as large sheets outside the task dialog DOM. > - The parent dialog's scroll lock cancels touch drags in those sheets. > - This pull request gives each mobile sheet its own modal scroll boundary. > - Users can scroll to an option, select it, and continue editing their task. ## Linked Issues or Issue Description **What happened?** An iPhone Safari user could not drag through the assignee or model list in the new-task composer. The list stayed at the top. A touch-enabled Chromium reproduction also showed canceled touchmove events and an unchanged scroll position. **Expected behavior** A finger drag scrolls the list. A tap selects an option. Closing the picker preserves the task draft and selected values. **Steps to reproduce** 1. At phone width, open New Task in a company with enough agents to overflow the picker. 2. Open Assignee and drag upward through the list. 3. Select a Codex agent, open Codex options, choose Custom, and open the model selector. 4. Repeat the drag with enough models to overflow the available viewport. **Paperclip version or commit** Reproduced on `e5bf9d49a`. The fix is rebased onto `b3eb03fcb`. **Deployment mode** Local development in an isolated test instance. The user reported iPhone Safari. Browser automation uses native Chromium touch input at phone dimensions; a physical iPhone was not available. Related: #14250 introduced the large mobile entity picker sheets. Duplicate search found no existing fix for their touch scroll boundary. ## What Changed - Use modal Radix popovers for mobile entity sheets. Their lists can scroll while the background stays locked. - Suppress opening when Radix restores trigger focus after dismissal. Escape and outside taps now close the picker without reopening it. - Add a failing-before/fixed-after regression for a mobile sheet inside a parent dialog. - Add a browser test for native touch scrolling in both lists, tap selection, draft retention, close-button/Escape/outside dismissal, and desktop mouse/keyboard selection. - Document the browser test command. ## Verification - Passed `pnpm -r typecheck`, `pnpm build`, and `pnpm check:token-gates`. - Passed 41 focused tests for `InlineEntitySelector` and `NewIssueDialog`. - Passed the new browser test against a disposable instance running this worktree. It uses 22 real fixture agents and a fixed 24-model catalog. It covers 390×844 and 390×430 phone viewports and a 1280×900 desktop viewport. - Inspected the rendered new-task form and both selectors. Selections returned to the draft with its title intact. - `pnpm test:run` was attempted and stopped after confirming that local database suites were skipping because macOS has exhausted its system semaphore limit (`initdb`: `could not create semaphores: No space left on device`). It did not complete locally. The complete test matrix passed in Linux CI, including all 71 tool-gateway tests. - All 150 browser tests passed in CI, including the new native-touch regression. - Greptile reviewed the latest commit at 5/5. Its desktop focus finding is fixed and the review thread is resolved. All checks on `2ab22727ca768b4134c9c840bbc8e4c80d7da674` are complete: 53 passed, 2 intentionally skipped Storybook deployment checks, and the Snyk status passed. ## Risks Low risk. Mobile selectors now own focus and scroll isolation. This changes their modal behavior, so the tests cover nested dismissal and focus return. Desktop selectors remain non-modal. No schema, API, or dependency changes. ## Model Used OpenAI Codex (GPT-6), with reasoning, repository inspection, code execution, and browser testing. The exact backend deployment ID 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 - [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>	2026-09-29 13:37:03 -05:00
DottaandPaperclip	b3eb03fcba	fix(sentry): preserve run failure stacks and diagnostic context (#14585) ... Builds on merged #14575 and targets `master`. ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Operators use optional Sentry reports to diagnose failed agent runs. > - The shared reporter currently converts each saved error message into a new exception. > - This loses the original stack and cause chain, and omits available adapter and provider diagnostics. > - This pull request selects, redacts, and bounds useful diagnostics before it sends them to Sentry. > - Operators can locate the failing operation and correlate upstream failures without copying arbitrary run data. ## Linked Issues or Issue Description Refs #14573. That change preserves ACP provider failure details in the run record. Refs #14575. That merged PR adds recorded exit codes and signals. This PR covers stacks, causes, and structured diagnostics without duplicating those fields. A thrown setup or adapter error currently appears in Sentry with the reporter's stack. A saved provider failure can contain useful details that never reach the Sentry event. Both cases use the same shared reporting path. ## What Changed - Pass caught setup and execution exceptions, and structured adapter error metadata, through successful terminal status transitions. - Snapshot a fixed set of execution, adapter, provider, and exception fields. Preserve up to four exceptions in the cause chain. - Remove registered secret values, declared runtime environment credentials, unknown inherited environment values and encoded credential forms, and credential patterns before truncation. Mark truncated fields and bound provider details and stacks. - Rebuild sanitized Sentry exceptions with the original stacks and causes. Use an adapter stack preview when available. Omit a fabricated reporting stack when the source has no stack. - Keep contexts local to each event. Preserve the optional DSN gate and existing error-code/adapter fingerprint. - Document the fields, limits, and omitted data. - Settle leftover chat fixture outbox rows only after assertions and worker shutdown, so subsequent tests cannot claim earlier cases’ pending actions or provider I/O. This fixes the CI shard contamination exposed during verification; production chat behavior and test timeouts are unchanged. ## Verification - `pnpm -r typecheck` passed. - `pnpm build` passed. - 86 focused diagnostic, Sentry, and startup tests passed after the rebase, including the real `@sentry/node@10.71.0` SDK with an in-memory transport. - Tests cover cause chains, HTTP status and request IDs, long provider details, secret redaction, failed secret resolution, cyclic causes, size limits, and context isolation. - Five targeted heartbeat integration cases passed, including thrown and returned errors containing an opaque environment-bound credential. The earlier full heartbeat integration run also passed its integration cases. - The focused suite also passed with an unknown inherited environment value set to `1`; reporting tests use a controlled environment and separately verify short-secret redaction. - Before the fixture cleanup, the full chat shard reproduced the CI Telegram timeout at `pending recovery before restart` (354 passed, 1 failed). After the cleanup, the same shard passed all 355 tests. No timeout or production behavior changed. - Latest-head CI (`0ae9e70df321d66dda025c3a7ba4787e169e0a4f`) passed typecheck, build, all 12 server shards, all 3 chat shards, workspace and serialized suites, browser shards, Runner checks, canary validation, the real Sentry SDK contract, and security checks. Required `ci / verify` and `ci / e2e` passed. - The branch has been rebased onto `master` after #14575 merged. All 55 applicable checks passed on this head (2 unrelated Storybook checks skipped). Greptile re-reviewed the current 11-file diff at 5/5 with no findings or unresolved review threads. - Local monolithic full-suite attempts were interrupted to apply fixes; full-suite success is not claimed from those runs. ## Risks - Error messages and stacks can contain credentials. The reporter uses existing redactors and the run's encrypted secret registry, reads only known fields, and skips capture if registered-secret resolution fails. - Unknown environment values remain private by default. Unrecognized short values can mask benign matches; known public settings are explicitly allowed. - Diagnostic text is bounded and can be truncated. Truncation is explicit. Data discarded upstream cannot be recovered. - These additional fields go to the operator's configured Sentry endpoint. Arbitrary request/response objects, headers, configuration, prompts, and stdout/stderr are not copied. ## Model Used OpenAI GPT-6 through Codex, with tool use and code execution. The exact serving model ID, context window, and configured reasoning level 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 - [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>	2026-09-29 12:54:12 -05:00
DottaandPaperclip	2de43fc909	fix(issues): keep agent mentions as context and defer personal app authorization (#14577) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Each task has one assignee. Explicit assignment and review requests select who should act. > - An agent mention started another agent on a task it did not own. Native attachment staging then rejected that run. > - Allowing that run through startup could also let two agents work on the same task. > - Mentions should identify relevant context. They should not start work or forward comments to other tasks. > - A personal app installed on a shared agent must also wait until tool use to resolve the current user's grant. > - This pull request removes mention dispatch and keeps missing personal app credentials from blocking startup. ## Linked Issues or Issue Description **What happened?** A native agent mentioned on another agent's task failed with `paperclip_runner_attachment_staging_not_authorized`. The source task could already be complete. A nearby optional-app warning was a separate problem: personal app tools were excluded when their shared health state required attention. **Expected behavior** An agent mention is context only. It does not wake the agent, take ownership, or copy a comment onto another task. Normal feedback still reaches the assignee. Assignment and explicit review requests still dispatch work. An unavailable personal app does not block startup or produce a startup warning. Tool use requests the current user's authorization and never uses another user's grant. **Steps to reproduce** 1. Assign a task to agent A. Post a comment that mentions agent B, including a comment that closes A's task or references B's child task. 2. Confirm the comment retains its agent link and B receives no run or deferred wake. A can still receive normal feedback. 3. Install an active personal MCP connection on B. Give only Alice a grant and leave shared health at `error`. 4. Explicitly assign work to B for another user. Confirm it can finish without using the app. 5. Ask B to use the app. Confirm its tool call shows an inline connection request for the current user. Related work: Refs #11144. This change uses the existing execution-time personal grant resolution. ## What Changed - Remove mention dispatch from standalone comments and issue updates. Remove implicit forwarding of parent comments to a mentioned worker's child task. - Ignore new requests with the legacy mention wake reason before creating a run or deferred request. Preserve already accepted queue entries, which can combine assignments and feedback with a later mention. - Remove the native mention admission, staging, and finalization exceptions from this PR. Native task ownership checks remain intact. - Keep active, installed personal app tools available despite shared health errors. Remove optional-app startup warnings. Tool execution retains the current user's grant and policy checks. - Update agent instructions and product/API docs. Refresh generated capability source anchors. ## Verification - Red: comment-route regressions reproduced extra agent wakes and child comment forwarding. A separate regression proved that cancelling by the last coalesced reason could drop an accepted assignment. - Green: the targeted route, wake queue, heartbeat, workspace, responsible-user, MCP discovery, and HTTP gateway suites passed. The final queue and heartbeat rerun passed 104 tests, the restored queue adapter passed 56, and both comment-route suites passed 135. These include accepted assignment preservation, rejection of new mention requests, and normal assignee feedback. - `pnpm -r typecheck` and `pnpm build` passed locally. The full local `pnpm test:run` attempt was interrupted for review/CI fixes, so it is not claimed as a completed local pass. It exposed a cleanup timing race in the concurrent-mention assertion, now fixed and verified across 10 repetitions. CI also exposed an obsolete test waiting for the removed mention lookup; it was reproduced and fixed, then both comment suites passed. Final full-suite verification is through CI. - Final head `bd9ea4cb05a8f081c54e017760a8999f9ea6ef44`: 54 checks passed, 2 Storybook checks intentionally skipped; no pending or failing checks. Full CI includes general and serialized suites, all 8 browser shards, runner verification, typecheck, build, and canary dry run. Greptile is 5/5 on this exact commit, with no unresolved findings. - One unchanged Cursor adapter test hit its 10-second CI timeout. All 5 tests in that file passed locally; one retry of its CI shard passed all 674 tests (3 skipped). The aggregate verification gate then passed. No code or timeout was changed for that retry. - Live browser check: inserted a structured mention with the picker on a human-owned task. The saved link remained visible. Database checks found zero new runs and zero wake requests. - Live Codex runner check: explicitly assigned that task with the unavailable personal app attached. The run succeeded and committed completion without using the app or creating a connection card. - Live browser follow-up: asked the assignee to call PostHog and mentioned another enabled agent as context. Only the assignee ran. It succeeded and displayed the existing inline connection card. Only Alice's grant existed; the run belonged to a different user. - The HTTP regression covers tool discovery with no provider calls or connection cards, first use returning the current user's authorization request, and successful retry after that user's grant exists. - App checks use an isolated local fixture and a fake MCP provider. They do not use production app credentials. ## Risks - Intentional behavior change: workflows that used mentions to wake agents must use assignment, a bounded child task, or an explicit review request. - Already accepted queue entries retain their prior rules. An old entry can combine assignment or feedback with a later mention; its last reason cannot safely identify mention-only work. New mention requests create no run or deferred wake. - Personal apps with a shared health error remain discoverable. Actual tool use still requires the responsible user's grant and existing policy gates. - No database migration or public API schema change. ## Model Used - OpenAI GPT-6 through Codex, with reasoning, repository tools, code execution, and browser testing. The exact serving model ID and context-window size are not exposed in this session. - Live native-run verification used `gpt-6-astra` through the Codex provider. ## 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>	2026-09-29 12:49:01 -05:00
Devin FoleyandPaperclip	e5bf9d49a5	fix(sentry): retain recorded process exit details (#14575) ... ## Thinking Path > - Paperclip manages agents and records their task runs. > - Operators can enable Sentry reports for terminal run failures. > - A failed adapter can leave only the generic message “Adapter failed.” > - The run already records a process exit code and signal, but the report omits them. > - This pull request carries those two values through a strict capture boundary. > - Operators can distinguish a nonzero exit from signal termination when the message is generic. ## Linked Issues or Issue Description **What happened?** A failed run can store `exitCode: 1` or `signal: "SIGTERM"` while its Sentry event contains only `adapter_failed` and “Adapter failed.” The existing reporter drops both recorded fields. This occurs on the current master reporting path. **Expected behavior** The opt-in report preserves bounded process exit evidence without exporting adapter output or changing run behavior. **Steps to reproduce** Enable the backend Sentry DSN and report a failed run whose message is “Adapter failed” and whose stored signal is `SIGTERM`. Before this change, the event has no signal field. After this change, `run_failure.signal` is `SIGTERM` and the existing fingerprint stays the same. Related: #12105 and #8222 describe missing adapter/HTTP failure details. #13152 changes terminal-result cleanup classification, and #12886 adds process-failure classification and runtime URL checks. None forwards these stored fields through the Sentry reporter. This change does not resolve those broader issues. ## What Changed - Forward the stored exit code and signal from the terminal run reporter. - Accept only signed 32-bit integer exit codes; use `null` for missing or malformed values. - Accept only the reporting host's Node signal constants; use `null` for missing values and `unknown` for unrecognized values. - Keep the added fields in event-local context, outside tags and fingerprints. - Cover database-backed reporting, malformed input, privacy, and isolation through the real Sentry SDK. - Document the fields and their limits. - Give the dedicated Sentry job the normal PR dependency-resolution fallback, with lifecycle scripts disabled on every install and the required real-SDK test retained. ## Verification - Before the change: 16 report-shape/exit-field assertions failed in the focused capture suite. - After the change: 91 focused Sentry, DSN, and database-backed reporting tests passed. The real SDK uses an in-memory transport. - Final real-SDK test also passed with malformed metadata; it verifies that arbitrary signal text is absent from captured events and unrelated errors inherit no run context. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - `pnpm test:run`: exited nonzero after 718 general-server files: 14,085 tests passed, 14 failed, 86 skipped. Thirteen skill-service/cache failures reproduce on the unchanged base commit on this macOS host. One comment-wake test timed out; the complete 29-test suite passes on the unchanged base and in final-head Linux CI. Local isolated rechecks skipped because embedded PostgreSQL could not start; these are not counted as passes. The remaining local workspace/serialized lanes did not run after the failing first lane; all CI lanes passed. - Initial Sentry CI failed before tests with `ERR_PNPM_LOCKFILE_CONFIG_MISMATCH`. Its install step lacked the normal PR fallback. The repaired real-SDK job passed. Security review then requested disabling lifecycle scripts for resolved dependencies; every install now uses `--ignore-scripts`. A fresh isolated checkout passed the exact script-disabled fallback and real-SDK contract. Final-head real-SDK CI and the security scan passed. - GitHub CI: all 54 checks passed on `37e0a836e50660f7753d367bcf5a4959eaf89b90`, including required `ci / verify` and `ci / e2e`; two unrelated checks skipped. - Greptile: 5/5 on that commit. No unresolved review comments. - Merge status: conflict-free; required CODEOWNER approval for the workflow change is still pending. ## Risks Low risk: this only adds two validated fields to existing opt-in error reports. It changes no database schema, run status, retry, fingerprint, or suppression rule. Process output and adapter result payloads remain excluded. A recorded signal does not identify its sender or prove an out-of-memory kill. Missing exit evidence stays unknown; this change does not establish the cause of a historical generic adapter failure. ## Model Used OpenAI GPT-6 (Codex), with reasoning, terminal tools, and code execution. The context window size is 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 references) - [x] My branch name describes the change and contains no internal ticket id or instance-derived details - [x] I have run focused tests locally and they pass; broader validation is recorded 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> canary/v2026.929.0-canary.6	2026-09-29 12:29:17 -05:00
github-actions[bot]andlockfile-bot	3166e93a7e	chore(lockfile): refresh pnpm-lock.yaml (#14461) ... Auto-generated lockfile refresh after dependencies changed on master. This PR only updates pnpm-lock.yaml. Co-authored-by: lockfile-bot <lockfile-bot@users.noreply.github.com>	2026-09-29 11:19:53 -05:00
DottaandPaperclip	29c8fb0b66	fix(composer): match effort options to adapter execution (#14576) ... ## Thinking Path > - Paperclip is an open source control plane for AI agents. > - The issue composer lets operators choose a model and effort for a task message. > - The picker must show only effort levels that the selected adapter can execute. > - The Codex Runner fix in #14568 exposed similar gaps in other adapters. > - Grok hid a supported control. Claude used one range for all models. Kimi ACP showed a control that it ignored. > - This pull request aligns the picker with each adapter and adds matching stories. > - Operators can see and save supported effort settings without false controls. ## Linked Issues or Issue Description **What happened?** The composer hid Grok effort. It showed an incomplete effort range for some Claude models and a slider for Claude Haiku. It showed Kimi effort on the default ACP engine even though that engine drops the value. **Expected behavior** The composer should show only effort levels supported by the selected model and execution engine. Grok effort should reach its adapter as `reasoningEffort`. **Steps to reproduce** 1. Select a Grok agent and `grok-4.7`. Observe that the picker has no effort slider. 2. Select Claude Sonnet 5 or Haiku 4.5. Observe the generic low to high slider. 3. Select a Kimi agent on ACP. Observe the slider even though ACP ignores effort. Related fix: #14568. ## What Changed - Use Claude and Grok adapter capability helpers in the production picker and Storybook preview. - Map Grok composer effort to `reasoningEffort` in per-message overrides. - Show Kimi effort only when its agent uses the CLI engine, including when it runs its default model. - Show Grok effort when it runs its default model. - Add design and production stories for Claude, Grok, and Kimi ACP and CLI cases. - Document the picker rules and add focused tests. ## Verification - `pnpm --filter @paperclipai/ui exec vitest run src/components/task-chat/composer-run-settings.test.ts` - `pnpm --filter @paperclipai/ui typecheck` - `pnpm check:token-gates` - `pnpm -r typecheck`, `pnpm test:run`, and `pnpm build` passed on the first commit. - The focused test, UI typecheck, token check, and Storybook build passed again after the review fixes. - Browser smoke: production stories show effort for default Grok and CLI Kimi, and hide it for ACP Kimi and Claude Haiku. ## Risks - Existing Kimi ACP agents lose a slider that could not change execution. Their stored effort override remains unchanged until the next edit. - Claude and Grok ranges follow their adapter catalogs. A provider can change model support before the catalog is updated. ## Model Used OpenAI Codex based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Reasoning, tool use, and code execution were used. ## 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> canary/v2026.929.0-canary.5	2026-09-29 10:34:29 -05:00
DottaandPaperclip	83076d7e7c	feat: return completed handoffs to Agent Chat (#14408) ... Return completed Agent Chat handoffs through a durable outbox and scope each generated update to its supplied tasks. Add recovery, browser delivery, result access, and calibrated quality coverage. Validated with two consecutive ten-case Claude/Codex campaigns, all CI checks, and a 5/5 review. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-29 10:25:52 -05:00
DottaandPaperclip	3b4b270650	fix(adapters): preserve ACP terminal failure diagnostics (#14573) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The shared ACP adapter engine records agent failures for operators. > - ACP providers can report a failure category, title, and detailed cause. > - Our patch kept only the category in the saved error, so an operator could not diagnose a failure when tracing was off. > - This pull request preserves redacted provider diagnostics in the run error, transcript, and structured run result. > - Operators can now inspect the provider message and any supplied request ID or stack trace after the run ends. ## Linked Issues or Issue Description Refs #13889 (the diagnostic gap; this PR does not update the bundled Claude version). Refs #14484 (related model-refusal classification; this PR retains diagnostics for all terminal failure categories). **What happened?** An ACP turn failed with only `ACP agent reported a terminal service failure.` The provider's title and details were available in memory but absent from the saved error and transcript. **Expected behavior** The run retains useful provider diagnostics even when raw tracing is disabled. Credentials remain redacted. A size limit must report truncation instead of silently removing the cause. **Steps to reproduce** 1. Run an ACP agent that returns an error-severity typed session failure. 2. Include an HTTP error, request ID, and stack text in its title and details. 3. Inspect the failed run with tracing disabled. Before this change, only the category survives. ## What Changed - Both pinned ACPX patches pass complete error text to the in-memory callback, so redaction happens before truncation. - The shared engine retains the sanitized category, title, and details in `resultJson.terminalSessionFailure` and includes the text in the run error and error transcript. - Diagnostics redact configured environment values even under arbitrary names, unknown launch-environment values, connection URL passwords, run credentials, and common credential syntax. Known boolean settings remain readable, while credential values are redacted even when embedded in other text. Diagnostics remove control characters and invalid Unicode. - Title and detail limits keep escaped transcript JSON below the server's chunk limit. Truncated fields include an omission count. The safe run-result projection preserves a byte-bounded diagnostic preview when the result exceeds its byte budget, with an explicit pointer to the full adapter-bounded run error and transcript. - The existing UI and CLI display the error. Diagnostics do not become assistant output. Issue continuation summaries and session-compaction prompts receive only the generic category, preventing provider text from becoming handoff instructions. Existing quota classification, warnings, timeout precedence, and control-channel failure precedence remain in place. - Regression tests cover real ACP child processes with both pinned versions in one-shot and persistent modes, credential redaction, request IDs after the old 4 KiB cutoff, transcript parsing, storage bounds, and database retrieval of oversized multibyte diagnostics. ## Verification - Full CI on `20ad4f5f1f66c46d2c260e6ad0339cbea607b4cf`: **54 passed, 2 intentionally skipped, no pending or failing checks**. Includes typechecking, build, all Vitest shards, Runner checks, browser E2E, and the canary packaging/public-install dry run. - Greptile: **5/5** on this commit. Superagent security scan passes. All review threads are resolved. - Local verification passed: shared ACP engine suite (395 tests); real Claude ACP child-process and diagnostic regressions across both pinned runtimes and both execution modes; run retrieval and model-handoff regressions (59 tests); ACPX patch packaging (16 tests); full typecheck and build. Affected package typechecks and focused tests were rerun after review fixes. - The broad local `pnpm test:run` was stopped after review edits made its cached imports stale. Fresh targeted runs pass, including both affected server suites. Cold-build import failures were also rerun after dependency builds: chat integration (1,063 tests) and tool access (351 tests) pass. The final commit's complete CI matrix is green. ## Risks - Provider diagnostic text is untrusted. This change retains more of it in company-scoped run records. Redaction and size bounds apply before persistence. - Diagnostics are limited to fields the provider supplies. Old runs cannot recover discarded error text. - No schema migration, recovery-policy change, or new Telemetry or OpenTelemetry export. ## Model Used - OpenAI GPT-6 through Codex, with reasoning, repository inspection, code editing, and test execution. The exact serving model ID 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 - [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>	2026-09-29 10:18:30 -05:00
3 people	da887ea3e9	fix(runner): honor Codex effort selected in composer (#14568) ... ## Thinking Path > - Paperclip manages AI agents that work on assigned tasks. > - The task composer lets a person choose an assignee, model, and effort for the next run. > - A Paperclip Runner agent can use Codex as its provider. > - The composer hid Codex effort for that agent because it checked only the older Codex adapter. > - The native Runner input also did not carry an effort choice to Codex. > - This pull request carries the chosen effort from the composer to each Codex turn. > - People can now select a supported effort and get the effort they selected. ## Linked Issues or Issue Description Refs #14322 **What happened?** The composer showed a model but no effort slider when the assignee used Paperclip Runner with the Codex provider. A task-level model override also did not reach the native Runner input. **Expected behavior** The composer shows effort choices for a known Codex model. The next native Codex turn uses the selected model and effort. **Steps to reproduce** 1. Open a task composer. 2. Select an agent that uses Paperclip Runner with the Codex provider. 3. Select a known Codex model such as `gpt-6-astra`. 4. Open the assignee and model picker. The effort slider is missing before this change. ## What Changed - Show known Codex effort levels for Paperclip Runner Codex assignees. - Save the task effort override in the native run input and send it to Codex on each turn. - Apply the task's merged model and effort overrides when the native run starts. - Apply a task model override for OpenCode Runner without changing the agent's provider. - Add Runner effort tests and desktop and mobile Storybook cases. ## Verification - `pnpm -r typecheck` passed. - `pnpm build` passed. - `pnpm build-storybook` passed. - `pnpm check:token-gates` passed. - Focused UI, server, Runner contract, and Codex driver tests passed. - The full CI test matrix, build, typecheck, and canary dry run passed on the latest head. ## Risks - Native Runner inputs add an optional Codex effort field to the current v5 input. Older inputs keep their previous behavior. - A known model rejects an effort that its catalog does not support. Unknown models do not show a slider. > This fixes an existing composer bug. I checked `ROADMAP.md`; it does not describe this bug as planned work. ## Model Used OpenAI Codex, GPT-6. The exact deployment ID and context window are not exposed in this session. The model used reasoning, code execution, and repository tools. ## 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> Co-authored-by: OpenAI GPT-6 Astra <noreply@openai.com>	2026-09-29 09:41:05 -05:00
DottaandPaperclip	3f4f8b37ba	fix: grade Codex clarification and refusal outcomes from evidence (#14570) ... Grade clarification lists, obsolete unstarted wakes, and refusal cancellation from persisted evidence. Preserve execution and ownership assertions, add boundary regressions, and version the affected eval definitions. Co-Authored-By: Paperclip <noreply@paperclip.ing> canary/v2026.929.0-canary.3	2026-09-29 09:40:55 -05:00
DottaandPaperclip	7636966452	fix(inbox): keep other users’ failed runs out of Mine (#14572) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The Mine inbox shows work that needs the current user. > - Failed-run rows used the latest run for every agent in the company. > - A failure from another user therefore appeared in Mine and its badge. > - Run list responses also omitted the responsible user needed to filter these rows. > - This pull request uses run ownership for personal failure routing. > - Users see their own failures and can still inspect company failures in All. ## Linked Issues or Issue Description **What happened?** An agent run started for one user failed. Its row and failure badge appeared in another user's Mine inbox. **Expected behavior** Mine and its badge include failed runs for the current responsible user. Other users' failures remain available in All and run details. **Steps to reproduce** 1. Use a company with two human users. 2. Create a failed or timed-out run attributed to the first user. 3. Open Mine as the second user. Before this fix, the failed run appears there and increases the badge. **Paperclip version or commit** Reproduced in regression tests on master at `24beb0057`. **Deployment mode** Authenticated deployment with multiple users. Tests also cover the local single-user board. Related prior work: #933 addressed inbox dismissal and badge consistency. No duplicate ownership fix was found. ## What Changed - Return `responsibleUserId` in normal and summary run lists. - Share one ownership rule across both inbox versions and client/server badges. - Select the latest run per agent before applying the ownership filter. This prevents old failures from resurfacing on shared agents. - Keep unattributed historical failures in the local board's Mine view. Hide them from authenticated users with no matching owner. - Keep company health alerts outside the personal badge, consistent with the client. - Document the routing contract and add page, badge, and database regression coverage. ## Verification - Red: the new badge cases failed with three company failures instead of one personal failure; eight Mine page cases failed across both inbox versions. - Green: 113 focused tests pass in `ui/src/lib/inbox.test.ts`, `ui/src/pages/Inbox.test.tsx`, `server/src/__tests__/heartbeat-list.test.ts`, and `server/src/__tests__/inbox-dismissals.test.ts`. - `pnpm check:token-gates` passes. - Agent calls on behalf of a user have two additional red-to-green API regressions. - Full `pnpm -r typecheck` and `pnpm build` pass. Server typecheck also passes after the agent-call fix. - All CI test shards and browser tests pass on `243bfa681`. The duplicate local `pnpm test:run` was stopped after the CI test lanes completed; it did not finish locally. ## Risks - Authenticated users no longer receive unattributed legacy failures in Mine. Those failures remain visible in All. - The server badge no longer counts company health alerts, matching the existing client badge. - No migration, run state, retry behavior, or company access rules change. ## Model Used - OpenAI GPT-6 through Codex, with reasoning, terminal execution, and browser tools. The exact deployment variant 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 - [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>	2026-09-29 09:40:21 -05:00
DottaandPaperclip	24beb00575	feat(runner): add rich ACP transport and durable interaction foundation (#14430) ... Add shared rich ACP transport, durable questions and permissions, verified provider packaging, and bounded activity and plan presentation. Keep Cursor, Copilot, and Pi pending their separate provider qualification. Persist interaction settlement before publication, fence failed writes until fresh recovery, and preserve owned-process cleanup. Incorporate reviewed mainline integration with extended harness coverage. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-29 08:56:21 -05:00
DottaandPaperclip	c9b93d7e8c	fix: preserve terminal task owners during release (#14561) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Tasks record an assigned owner and separate checkout and execution locks. > - A completed task must retain its owner after execution ends. > - The release endpoint currently clears that owner when it clears the locks. > - This pull request preserves the assignee of Done and Cancelled tasks during release. > - Unfinished tasks keep the existing relinquishment behavior. > - The benefit is stable task attribution without retaining execution locks. ## Linked Issues or Issue Description **What happened?** An agent completed an assigned task, then called the release endpoint. The task stayed Done, but its assignee became null. The same defect affects Cancelled tasks. It caused the legacy Claude clarification/reuse and multiple-repository handoff E2E assertions to fail. **Expected behavior** Release must clear execution locks on terminal tasks and preserve their assignee, final status, and disposition timestamps. Release of unfinished tasks must still clear the agent assignee. Only In Progress work returns to Todo. **Steps to reproduce** 1. Create an assigned task with checkout and execution locks. 2. Complete or cancel the task. 3. Call `POST /api/issues/:id/release` as the assigned agent. 4. Read the saved task. Before this fix, its assignee is null. **Paperclip version or commit** Reproduced on master commit `d172197117a14b80a1eb2d2835a0e7cce2679656`. **Deployment mode** Local tests against real PostgreSQL through the production issue routes and services. This is a core lifecycle defect, independent of the agent adapter. Refs: #11689, #6899, #7769. These are related open release proposals. This is an independent fix limited to terminal task ownership. It does not include timer scheduling changes. ## What Changed - Preserve the current assignee when releasing Done or Cancelled tasks. - Keep all execution-lock cleanup and existing unfinished-task behavior. - Cover all seven task statuses through the release API and read back saved state. - Check disposition timestamps, activity attribution, and repeated board cleanup. - Update the API contract, agent reference, and CLI help. ## Verification - Red commit `ddaabb754`: the two terminal-owner regressions failed with `assigneeAgentId: null`; 12 other route tests passed. - Green: all 14 route tests pass, plus the existing successor-checkout race test (15 selected tests total). - Command: `pnpm --filter @paperclipai/server exec vitest run src/__tests__/issue-stale-execution-lock-routes.test.ts src/__tests__/issues-service.test.ts -t 'stale issue execution lock routes|does not let stale release clobber a successor checkout lock'`. - The local host has exhausted its SysV semaphore pool. The red/green runs used the existing test-provider hook to start disposable Docker PostgreSQL 17 instances. Routes, services, migrations, and assertions were unchanged. No database tests in the selected set were skipped. The other 134 tests were excluded by the name filter. - Capability contract and inventory drift checks pass. - `pnpm build` and `pnpm -r typecheck` pass. - The local `pnpm test:run` was interrupted after environment failures while the complete sharded CI suite ran in parallel: native PostgreSQL bootstrap fails under the host semaphore limit, and the large Git fixture hits macOS `ENAMETOOLONG`. A focused rerun confirmed these happen before the relevant assertions. The interrupted local run is not counted as a full pass. - Greptile completed on `1caeeb827e9cb658ddb71f16c2421ec20f80634e` with **5/5**, a successful check, and no review threads. - All CI gates pass on the current head: typecheck, build, general and serialized tests, Runner checks, browser E2E, release packaging, and security checks. Server shard 11 passed on one targeted retry; the first attempt had 836 passing tests but an unhandled workspace-runtime startup rejection caused by an existing timing window. All other successful jobs were reused. - No paid provider evaluations were run. ## Risks - A caller that used release to erase ownership from terminal work will now retain that owner. An explicit assignment update or the board force-release option with `clearAssignee=true` can still clear it. - No schema or migration changes. The transaction, company access, assignee/run checks, and activity log remain in place. ## Model Used - OpenAI GPT-6 through Codex, with tool use, code execution, and test debugging. The session does not expose a more specific backend model version or context-window size. - OpenAI `gpt-6-luna` assisted with read-only test discovery and review. ## 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>	2026-09-29 08:47:33 -05:00
DottaandPaperclip	61b3fd57a6	fix(ui): recover gracefully during server restarts (#14560) ... Show a clear reconnecting state during server restarts and retry safe access checks every five seconds. Preserve open drafts, wait for initial startup readiness, and keep authentication failures separate from temporary outages. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-29 08:31:05 -05:00
DottaandFry	3ca196b0a6	feat(agents): persist agent files across tasks without revision history (#14420) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - An agent needs personal files across tasks and sessions. > - AGENTS.md is one file in that directory. Supporting files need the same persistence. > - The Instructions Editor and agent runs must share one current directory. > - Concurrent runs should apply only the files they change. The last sync of the same file wins. > - This pull request uses existing file transport and removes temporary copies after sync. > - Old instruction-only sessions keep their restore contract. New saves do not create revision history. ## Linked Issues or Issue Description Refs #14325. This replaces its revision-oriented design with persistent agent files. Keep #14325 unmerged. Transport prerequisite #14416 merged first at `d172197117a14b80a1eb2d2835a0e7cce2679656`. This PR now targets master and remains below 100 changed files. Related work: #4513 and #8798 cover instruction tooling. This change handles run synchronization, cross-task personal files, browser editing, and old-session restoration. ## What Changed - Keep one current directory per company and agent. Point AGENT_HOME at a temporary working copy for each active run. Keep task files and provider HOME separate. - Restore text, binary files, and nested folders through workspace transport. Exclude remote agent files from task Git snapshots with a self-ignoring file inside the reserved runtime directory; never write through repository-controlled Git metadata. - Collect after the provider and child processes have stopped. Keep resumable conversation state. - Apply changed and deleted files under the agent lock. The last sync wins for the same file. Unrelated concurrent changes survive. - Remove temporary copies after successful sync, rejected sync, and staging failure. Register ownership before copying so restart recovery can remove interrupted preparation. Retry transient synchronization up to three times. Preserve the original remote lease reference until deletion succeeds; restart cleanup never acquires a replacement sandbox. Do not create captured directories or a conflict-review queue for new runs. - Keep browser editing, stale-draft protection, and streaming binary downloads. Keep the instruction entry and text editor limited to 1 MiB. - Keep historical agent-folder sync failures on their affected runs instead of repeating them above current saved instructions. Preserve legacy candidate review and current browser-save errors. Avoid duplicate quota warnings while retaining separate sync failures when they describe a different problem. - Require target-scoped caller grants for peer instruction access, while preserving self edits, responsible-user checks, and protected-change consent. - Treat full storage as a nonblocking run warning, never an agent pause or run-admission failure. Restore already-over-quota saved folders so ordinary agent cleanup can recover; warn on each run until cleanup. The run detail view shows the warning. - Allow 256 MiB per file, 2 GiB per directory, and 100,000 entries. Hash large files as streams. Check editor-save quotas with metadata instead of hashing unrelated files. - Preserve old native inputs, instruction-only copies, paths, digests, and pending legacy candidates. Adopt old revision heads once. New writes do not append history rows. - Add idempotent migration 0287 and verify upgrades from the preview tables and receipts. - Add nine interactive stories under **Agents / Persistent files**, including automatic incoming edits, stale browser drafts, and storage-limit diagnostics. ## Verification - Merge candidate: `4f5390107ec6ffd80a76d1d2e85530e66f21d079`, after merging current master and the landed transport prerequisite. Integration required no manual conflict resolution; the feature remains 99 changed files. Full workspace typecheck, production build, token gates, and 715 focused tests passed on this merge candidate. Fresh Greptile review is 5/5 with no unresolved findings. All 55 checks passed, with four conditional skips, including the build, typecheck, browser E2E, and canary dry run. A single retry recovered four jobs interrupted by runner shutdowns; no source changes were required. - Historical-warning UI fix: all 6,834 UI tests across 640 files passed, including regression coverage for three old failures, legacy preserved edits, and warnings scoped to the affected run. Full workspace typecheck, production build, Storybook build, and token gates passed. Browser-verified Storybook playtests passed for Historical Failures After Successful Save, Storage Limit, and Full Storage Run Warning. - Review follow-ups at `4e20c9fb2`: all 18 focused tests passed, including external Git directories, linked worktrees, symlinks, hardlinks, and distinct I/O failures alongside storage warnings. Server and UI typechecks, token gates, and the production build passed. - Storage warning regressions at `0724f3012`: all 33 directory tests and all five heartbeat-list tests passed, with no skips in their successful runs. They cover repeated runs while full, an already-over-quota saved folder, cleanup, warnings retained after unrelated save failures, and bounded warnings in large result JSON. Server typecheck passed after the final warning fixes. - Full workspace typecheck, production build, and token gates passed during this follow-up. Product E2E harness: 631 tests passed across 52 files; harness typecheck passed. Earlier native session/context and directory/legacy collection suites passed 537 tests; Runner unit/transport suites passed 329 tests. - **Real E2E at `0724f3012` (before this follow-up):** legacy local Codex and native Daytona Codex each passed six tasks, one server restart, seven independent assertions, and cleanup verification. Both prove browser-to-agent edits, agent-to-browser edits, nested/binary restoration, per-file last-sync-wins, a successful run after an oversized save rejection, and cleanup clearing the warning. - Native local Codex also passed the six-task quota flow before the final warning-retention fixes. That pass began at `918d1ed02` while the bounded-result warning fix was being edited, so it is not claimed as exact-final-head evidence. Its final-head rerun failed during embedded PostgreSQL bootstrap before any provider run: the macOS host had 87,365 of 87,381 SysV semaphores occupied. No unrelated services or kernel limits were changed. - The final-source report intentionally records **2/3 cells passed**, preserving the blocked native-local attempt: `tests/runner-e2e/results/agent-files-quota-final-20260928-report/`. Earlier failed attempts and provenance notes remain under `tests/runner-e2e/results/agent-files-quota-final-20260928-input/` and the original campaign directories. - Daytona used immutable image `ghcr.io/paperclipai/paperclip-daytona-runner@sha256:5643f0d801417cae3581833a1a3bc6715b325e028602738d2652c44cac5dc6bf` and its exact Linux runner binary. Controller source is `0724f3012`; image source is recorded separately. - Legacy-session compatibility and all three ACP Stop/resume browser regressions passed on the prior validated feature head `169fab46d5af21caa2269b4c1b29b69c933a6951`. They assert the same provider session is retained and interrupted writes are not replayed. Migration upgrade tests also passed earlier. - Nine interactive stories are under **Agents / Persistent files**, including **Full Storage Run Warning**. Its playtest and visual browser inspection passed; the warning states that runs continue and the editor remains available. - Prior-head checks on `4e20c9fb2`: 55 passed, two conditional jobs skipped, no failures or pending checks. All eight browser E2E shards and their aggregate passed. Fresh Greptile review is 5/5 with no findings; all review threads are resolved, the security scan passed, and GitHub reports no merge conflicts. - The broad local follow-up test run was interrupted after host semaphore exhaustion affected isolated PostgreSQL instances. It also encountered the existing macOS long-path fixture failure and two timeout failures. This is not a claim that the full local suite passed. Logs are retained; focused storage/warning tests passed. ## Risks - A later sync can overwrite an earlier edit to the same file, including a saved browser edit. There is no text merge or retained version. This is the intended last-sync-wins policy. - A save that exceeds a storage limit is rejected and its temporary copy is discarded. The run itself continues normally, and later runs restore the last saved files with a warning until cleanup. Transient sync failures get bounded retries. An I/O failure partway through a sync can leave some files updated; a failed receipt does not claim whole-folder success. - Larger folders increase copy time, network traffic, and temporary disk usage. Active runs still need working copies. Terminal runs do not accumulate archives. Operators must provision disk for agents and configured concurrency; these limits are not company-wide quotas. - A restored old native session remains instruction-only until a fresh session starts. Its original conflict fence and existing pending candidates remain compatible. - Provider processes close at the collection boundary. Conversation resume remains available, but warm process reuse is lost. - Backups must include the instance filesystem and database. External bundles keep their existing behavior until explicitly moved to managed storage. ## Model Used OpenAI Codex, GPT-6 family. The session does not expose a more specific model ID or context-window size. Reasoning, code execution, and browser tools assisted this change. Real provider E2E uses `gpt-5.6-sol`. ## 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: Fry (Paperclip) <noreply@paperclip.ing>	2026-09-29 08:25:56 -05:00
DottaandPaperclip	d172197117	feat(storage): add plain directory sync with conflict preflight (#14416) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agent runs use workspace transport to restore and collect files. > - Some directories belong to the agent across tasks. > - Those directories need plain file transport without task Git state. > - A concurrent file edit must be detected before a merge changes any file. > - This pull request adds optional plain-directory sync and conflict preflight. > - Existing task workspace sync keeps its defaults. ## Linked Issues or Issue Description Refs #14325. This is the transport prerequisite for a replacement of its instruction revision design with current agent files. ## What Changed - Add an opt-in plain-directory transport mode to command and sandbox runtimes. - Add strict merge preflight for file edits, deletions, and directory changes. - Accept identical replay after an interrupted merge. Preserve competing changes. - Set the compiled OpenCode test executable to 0755, independent of the CI host’s file-creation mask. Preserve the original startup error if cleanup also fails. ## Verification - At `ced53ae532ce6966cad5a83575d83db61af98126`, all 212 targeted transport tests passed across workspace restore, remote managed runtime, SSH fixture, and execution-target sandbox suites. Adapter-utils typecheck passed. - Real isolated SSH retry fixture previously passed with `PAPERCLIP_ENABLE_DARWIN_SSH_ENV_LAB=1`; stale deleted files remain absent while gitignored binary bytes survive. - Dependent PR #14420 passed real native local, legacy local, and native Daytona persistence E2E at `169fab46d5af21caa2269b4c1b29b69c933a6951`, which includes all production transport changes through `ced53ae53`; the subsequent two commits only fix the OpenCode test fixture. Nine tasks, three server restarts, and all cleanup checks passed. - A hosted OpenCode fixture failed twice at `ced53ae53`. Reproduced the failure locally and in Linux with `umask 0002`: the compiler created a group-writable executable, correctly rejected by the qualified launch boundary. Explicit 0755 permissions fix the test without weakening the production guard. The focused test and non-root Linux reproduction now pass under that same mask. - Before rebase, head `69e97de0475d34aac5d532e559a405eaf015fd2b` includes the deterministic fixture permission fix and preserves original bootstrap diagnostics. All production transport code is unchanged since the 212-test validation. Fresh Greptile review is 5/5 on this exact head with no unresolved findings. All 54 current-head checks passed, with two conditional skips. The full CI run completed successfully, including the previously failing OpenCode runner shard. - Merge validation on rebased head `c509d79dd190c5cb00dc65edfde209097ff21465`: all five commits are patch-identical to the reviewed branch. All 54 checks passed with two conditional skips, and fresh Greptile review is 5/5 with no findings. One retry cleared an npm archive 404 and a Cursor fixture timeout. ## Risks - New behavior is opt-in. Existing task snapshot behavior retains its defaults. - Generic strict merge preflight remains opt-in. The dependent agent-folder feature rebases changed paths before applying them to provide per-file last-sync-wins; it does not create a conflict-review queue. - This change adds no database migration, dependency, or UI. ## Model Used OpenAI Codex, GPT-6 family. The session does not expose a more specific model ID or context-window size. Reasoning, code execution, and tool use assisted this change. Live provider validation used `gpt-5.6-sol`. ## 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>	2026-09-29 07:59:11 -05:00
Devin FoleyandPaperclip	53aad90b9e	fix: retry sandbox ACP input delivery after gateway failures (#14485) ... ## Thinking Path > - Paperclip coordinates agent work through execution adapters. > - Sandbox ACP sessions send ordered input through a remote file queue. > - A temporary provider 502 currently closes the session during an input upload. > - A lost response can occur after the sandbox has consumed the message, so a blind retry can duplicate input. > - This pull request retries gateway failures with the same sequence and drops consumed sequences at the receiver. > - The session can continue through a brief provider failure without repeating a tool call. ## Linked Issues or Issue Description **What happened?** A sandbox ACP run can fail with `ACP agent disconnected during request (connection_close, exit=null, signal=null)` when a provider input upload returns HTTP 502. The bridge destroys its local socket on the first failure and can discard the diagnostic before the proxy reads it. **Expected behavior** A temporary gateway failure should get a bounded retry. A lost response after successful delivery must not duplicate input or reorder later messages. Permanent failures must still close the session. **Steps to reproduce** 1. Run the real sandbox process bridge with an echo child and a local test runner. 2. Inject a provider 502 before preparation, after a chunk upload, or after final publication and consumption. 3. Send the next input message. Before this change, the connection closes instead of delivering it. Searched open and closed PRs for `ACP disconnect`, `bridge retry`, and `502 sandbox`. Related work: #13287 covers shutdown after bridge loss; #13793 covers large launch envelopes. This change covers ordered input delivery within a running legacy ACP session. ## What Changed - Retry input uploads up to three times for recognized Daytona and Cloudflare HTTP 502, 503, and 504 diagnostics, with 250 ms and 500 ms delays. - Give each upload separate temporary paths and discard already-consumed input sequences, including late publication from an earlier attempt. Clean failed attempts in the background without removing a published message or another attempt’s files. Cleanup cannot delay retries or shutdown. - Keep later input behind the retry. Stop queued input on permanent failure and flush a fixed diagnostic before closing the socket. Neither failure-diagnostic persistence nor shutdown-warning persistence can block teardown. - Add real-process regression tests for lost responses, late publication, retry exhaustion, immediate permanent failure, and diagnostic redaction. - Give accepted run-log file appends up to three seconds to drain before finalization computes the size, hash, and durable copy. Close the run handle to later appends. This waits only for file writes, independently of later DB progress or live-event persistence. If writes remain stalled, return null size/hash metadata and skip the final durable copy so the run can settle. Late writes cannot restart mirroring. - Preserve legacy comment attribution when final log size is unknown by reading existing entries within the unchanged 2 MB scan limit. Storage errors or a three-second read deadline return the evidence already read instead of failing the comment listing; pagination stops at the deadline. The deadline requests cancellation of the underlying local stream or S3 HEAD, GET, and response stream. A separate response timeout returns partial evidence even when filesystem I/O delays cancellation; late reads cannot append evidence or start another page. Each listing retains its existing batches of eight reads, without a shared admission cap that skips readable logs under contention. - Document the retry and log-finalization boundaries in the development guide. ## Verification - Final commit `347daa564b`: [Linux CI](https://github.com/paperclipai/paperclip/actions/runs/36506995168/attempts/2) passed. Greptile Apex review 13 scored this commit 5/5 with no new findings; all 12 review threads are resolved. - The final CI run initially hit a Cursor test timeout and four Discord credential-lock contention failures. All five cases passed in isolation. The two failed shards and their aggregate gate passed on retry without a code change. Those intermittent failures are not claimed fixed by this PR. - `pnpm --filter @paperclipai/adapter-utils typecheck` passed. - `pnpm exec vitest run packages/adapter-utils/src/execution-target-stdin-race.test.ts packages/adapter-utils/src/execution-target-sandbox.test.ts packages/adapter-utils/src/sandbox-callback-bridge.test.ts`: 262 tests passed on the final implementation, including 21 new regressions. The original three fault-injection cases failed before the fix. - The regressions cover failed and indefinitely stalled cleanup, Cloudflare gateway responses and retry exhaustion, permanent errors that must not retry, and teardown while failure logging remains indefinitely stalled. Seven Apex regression cases failed before the review fixes. Adapter-utils typecheck and build passed again after the final review change. - `pnpm exec vitest run server/src/services/run-log-store.test.ts server/src/services/run-log-store-cancellation.test.ts`: all 25 tests passed, including four new regressions that failed before the finalization fix. They cover delayed and failed appends, late-write admission, agreement between the local bytes/summary/durable copy, and a stalled append that exhausts the three-second budget. The timeout case verifies unknown metadata, no final upload, and no mirror restart after late completion. New cancellation tests use the real AWS SDK against a local HTTP server. They verify that stalled HEAD, GET, and response-body connections close on abort and that a subsequent read succeeds. Local range and already-aborted read cases also pass. - `pnpm exec vitest run server/src/__tests__/issues-service.test.ts -t 'readIssueCommentRunLogText|deriveIssueCommentRunLogAttribution'`: 14 targeted tests passed. The null-size reader case, both storage-error cases, the stalled-read case, and the cancellation/concurrent-listing cases failed before their fixes. The new regressions verify that timed-out reads are cancelled, subsequent listings recover, and two concurrent listings both retain their attribution markers. A read that ignores cancellation still returns partial evidence at three seconds and cannot resume pagination when it finishes; this regression failed before the response-timeout fix. - `pnpm --filter @paperclipai/server typecheck` and `pnpm --filter @paperclipai/server build` passed after the response-timeout change. - Full `pnpm -r typecheck` and `pnpm build` passed earlier in this PR; the affected packages were rechecked after review fixes. - Full local `pnpm test:run` failed in the general-server group: 511 files passed, 40 failed, and 158 were skipped. Failures include embedded PostgreSQL initialization, read-only cache directory renames, a macOS long-path fixture, and a workspace exposure assertion. The PostgreSQL, cache-permission, and long-path failures also reproduce with both changed implementation files restored to baseline commit `24c58e479a`. The exposure suite passes in isolation both on baseline and the fixed branch (28 passed, 3 skipped). CI runs the full suite on Linux. Later local test groups were not reached. - An earlier CI run hit the Telegram retry-timing failure fixed upstream in #14501. The branch includes that master fix. The selected recovery test passed against a fresh, migrated PostgreSQL 16 database. The embedded PostgreSQL runner is unavailable on this Mac; the isolated database was stopped and removed afterward. - No live agent turn was replayed. The tests use local child processes and injected provider failures. ## Risks Retries are restricted to recognized Daytona SDK and Cloudflare bridge gateway-error messages, which survive plugin RPC serialization. Other errors fail immediately. Temporary upload paths are now unique for all command-managed queue writes. Receiver sequence checks prevent duplicate input; retries do not restart an agent turn. Cleanup and failure logging are nonblocking and best effort; session teardown remains the final cleanup boundary. Log finalization now drains accepted local file writes for at most three seconds and ignores later appends on the closed run handle. A timeout leaves final size/hash unknown and skips the final durable upload; an existing partial mirror may remain available, but it is not claimed as a verified final snapshot. It does not wait for later DB progress or live-event persistence. Optional attribution keeps partial evidence when a read fails or times out. Cancellation closes S3 requests and response streams. Local filesystem I/O may finish after the caller deadline, but a late read cannot change the returned evidence or continue pagination. Later listings can retry after storage recovers. There is no schema, authentication, or permission change. Revert this commit to restore the previous behavior. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository inspection, code editing, 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 #` 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 and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally; targeted tests pass and full-suite limitations 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> canary/v2026.929.0-canary.1 nightly/v2026.929.0-nightly.0	2026-09-28 19:15:33 -07:00
Devin FoleyandPaperclip	ea371b9684	fix: retain project defaults in partial workspace overrides (#14502) ... ## Thinking Path > - Paperclip manages AI agents and their work. > - Project workspace policies define how isolated worktrees are set up. > - Tasks can override a branch without providing every setup field. > - The resolver currently replaces the entire project strategy with that partial override. > - Losing an explicit setup command can run the repository fallback script and block the task. > - This pull request keeps enabled project defaults when the task uses the same strategy type. ## Linked Issues or Issue Description **What happened?** A project uses `git_worktree` with `provisionCommand: "true"`. A task overrides only `baseRef`. The resolver drops the command. Worktree creation then invokes `scripts/provision-worktree.sh`, which can fail because its required setup is absent. **Expected behavior** A branch override keeps the project's provision, runtime provision, and teardown commands unless the task explicitly overrides them. A different strategy type must not inherit those commands. **Steps to reproduce** Configure the project with an enabled `git_worktree` strategy and `provisionCommand: "true"`. Give the task an isolated workspace with a `git_worktree` strategy and a different `baseRef`. Add a failing repository fallback provisioner. Before this change, worktree creation invokes that script. After this change, it uses the project's explicit command and succeeds. Related: #4968 concerns agent strategy and working-directory fallback. #13903 concerns gated API fields and reusable-workspace updates. #11091 concerns provision hooks on workspace reuse. None fixes partial task overrides discarding project defaults. ## What Changed - Merge a partial task strategy over the enabled project's strategy only when their types match. - Preserve explicit null values when parsing nullable strategy fields, so they can clear project values. - Keep explicit empty-string overrides and agent fallback behavior. - Exclude disabled project strategies and avoid an inherited branch template when a task pins an existing branch. - Add policy regression coverage and a real Git worktree test with a failing fallback script. - Document inheritance, explicit clearing, and no-op provisioning in the development guide. ## Verification - Policy regression: eight failures before the fix; all 41 policy tests pass after it. - Real worktree regression: passes and creates a worktree using the task's base branch without invoking the failing fallback provisioner. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - `pnpm test:run`: general-server phase completed with 13,873 passed, 86 skipped, and 14 failures in unchanged macOS skills-cache and Git long-path tests. The same failures reproduce on unmodified base code. The command stops at that phase, so no full local pass is claimed. - CI initially failed the existing Telegram subscription recovery test on a 15-second timeout. The separate fix and investigation are in #14501. A serialized job also lost its runner; GitHub reported lost communication, and that job was rerun without source changes. All 52 final-commit checks pass, with two intentional skips. Greptile is 5/5, with no unresolved comments or merge conflicts. The chat shard passed on one unchanged rerun. The timeout cause remains unproven; #14501 adds phase diagnostics for a recurrence. ## Risks Tasks that specify a partial strategy now retain the project's omitted fields, including setup and teardown hooks. This is the intended behavior change. Inheritance requires an enabled project policy and matching strategy types. Explicit task values still win. Null and empty commands restore existing runtime defaults; they do not guarantee that no script runs. Use `"true"` for an explicit no-op provision command. No migration, live configuration change, or task replay is included. ## Model Used OpenAI GPT-6 (Codex), with reasoning, terminal tools, and code execution. The context window size is 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 references) - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run focused tests locally and they pass; full-suite status is recorded 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> canary/v2026.929.0-canary.0	2026-09-28 18:09:20 -07:00
Devin FoleyandPaperclip	90119181e7	test: control Telegram subscription retry timing (#14501) ... ## Thinking Path > - Paperclip manages AI agents and their work. > - Telegram delivery recovers subscription changes after a restart. > - The recovery test leaves a failed action on the real one-second retry timer. > - A slow runner can cross that deadline before the test checks that no retry occurred. > - This pull request holds the fixture deadline until the explicit restart transition. > - The test still checks that recovery uses fresh provider options. ## Linked Issues or Issue Description **What happened?** The Telegram subscription recovery test expected one `setWebhook` request but saw two. A 1.5-second delay after the first failed request reproduces the failure. **Expected behavior** The test controls when the failed request becomes eligible for retry. Host speed does not change its result. **Steps to reproduce** Run the test named `retries an unknown subscription mutation after restart` with a 1.5-second delay after the first failed-action assertion. The old fixture retries too early. The updated fixture passes with the same delay. Related: #13952 fixes a separate Telegram fixture cleanup problem. This change addresses retry timing. ## What Changed - Set the stored retry deadline to 2099 before the pre-restart assertions. - Keep the existing explicit epoch deadline after restart and all provider request assertions. - Report the current test phase only when this test fails, to diagnose an observed intermittent CI timeout. - Leave production retry code unchanged. Temporary delay and per-step console tracing are not included. ## Verification - Delayed regression: failed before the change with two requests instead of one; passed after the change. - Focused Telegram durable private draft Stop group: 34 tests passed. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - Full local chat shard: 355 tests passed twice. - `pnpm test:run`: general-server phase completed with 13,861 passed, 86 skipped, and 14 failures in unchanged macOS skills-cache and Git long-path tests. The same failures reproduce on unmodified base code. The command stops at that phase, so no full local pass is claimed. - CI exposed a separate 15-second timeout. A diagnostic run passed all 355 shard tests, with the affected test completing in under one second. Its cause remains unproven. Normal step logging is removed; a failure-only phase report remains for a recurrence. The final commit also passes the 355-test chat shard, and Greptile rates it 5/5. All 52 final-commit checks pass, with two intentional skips. There are no unresolved review comments or merge conflicts. ## Risks Low risk. This changes only the fixture deadline. It does not disable a test, extend a timeout, or change production retry behavior. Existing assertions still verify the failed action, the pending state, restart recovery, and fresh provider options. The intermittent CI timeout is not claimed fixed; phase diagnostics narrow the next occurrence without changing the timeout. No documentation change is needed for a test fixture correction. ## Model Used OpenAI GPT-6 (Codex), with reasoning, terminal tools, and code execution. The context window size is 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 references) - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run focused tests locally and they pass; full-suite status is recorded above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes, or explained why none is needed - [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>	2026-09-28 18:08:57 -07:00
Devin FoleyandPaperclip	ad1f7e98ea	fix: retain diagnostic reasons for native runner failures (#14481) ... Retain bounded reasons for runner identity, harness recovery, and provider-pack read failures. Preserve existing ownership and cleanup proofs and compatibility with receipt-gated chat recovery. Verified with executor, recovery, diagnostic privacy, typecheck, build, and full CI checks. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-28 16:16:19 -07:00
Devin FoleyandPaperclip	4f6cf5b3ff	fix: prevent run identity locks from blocking audit checks (#14478) ... Use NO KEY UPDATE for identity locks so audit foreign-key checks can proceed while identity writers remain serialized. Preserve task-before-run ordering, company scoping, and foreign keys. Verified with PostgreSQL concurrency regressions, focused tests, typecheck, build, and full CI. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-09-28 16:16:05 -07:00
Devin FoleyandPaperclip	faf72cb1d5	fix: preserve company context across browser hot reload (#14482) ... ## Thinking Path > - Paperclip uses a shared company context for browser providers and consumers. > - Vite can load a new consumer module while an older provider is mounted. > - Recreating the context disconnects that consumer from the mounted provider. > - The consumer then reports a missing provider even though one is in its React ancestry. > - This change preserves the context object across development module refreshes. > - Bounded global error diagnostics distinguish development bundles and otherwise context-free promise rejections. ## Linked Issues or Issue Description **What happened?** A refreshed company consumer can throw `useCompany must be used within a CompanyProvider`. A real Vite and Chromium reproduction confirms that a retained provider and a refreshed consumer can hold different context objects. Global promise rejections also lack the bounded document state already attached to React boundary errors. **Expected behavior** A refreshed consumer should read the mounted provider. Error reports should identify the loaded bundle mode and bounded browser state while preserving monitoring opt-in, sign-out, and privacy behavior. **Steps to reproduce** Run `pnpm test:e2e:browser-context`. The isolated Vite fixture renders the real CompanyProvider, imports a new timestamped consumer module, and renders that consumer below the retained provider. The test fails before the context change and passes after it. The SDK regression invokes its real unhandled-rejection handler with an undefined reason. ## What Changed - Keep the React context object in Vite's per-module `hot.data`. Account values stay in React. - Add an isolated browser regression with mocked API responses and no live instance, discovered by the existing Chrome CI shards. - Add document-state diagnostics to global errors while preserving earlier boundary snapshots. - Tag events with development or production bundle mode and the type of an unhandled rejected value. - Document the new test command and diagnostic fields. ## Verification - Company context, browser context, and Sentry suites: 59 passed. - `pnpm test:e2e:browser-context`: passed in Chromium. The original context code fails the reproduction. - Real SDK tests preserve DSN/sign-out behavior and omit request context, breadcrumbs, and private DOM data. - `pnpm check:token-gates`: passed. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - Local `pnpm test:run` exited in the general-server phase: 13,819 passed, 86 skipped, 21 failures in unchanged filesystem and host-port suites. Cache permission and long-path failures also reproduce on the unmodified base; seven other failures involve local runtime port ownership. This is not a full local pass. - Full Linux CI and review passed at a0b5270662 (53 successful checks; two conditional checks skipped). Three jobs interrupted by a runner shutdown passed on rerun. Greptile is 5/5 with no unresolved comments. The real browser regression also passed in the normal Chrome CI shard (3.2 seconds). - Code-owner approval remains required because the dedicated test command changes `package.json`. - Scanned the diff and PR text for credentials and private data before pushing. ## Risks The context cache applies only to development hot reload. It retains the context object, not account state. Production continues to create an ordinary React context. The diagnostic hook adds only bounded state and type values, preserves boundary snapshots, and returns the original event if enrichment fails. It does not suppress errors or restore raw breadcrumbs. The global rejection diagnostics do not identify the promise's originating operation by themselves. ## Model Used OpenAI GPT-6 through Codex, with repository inspection, code editing, and command execution. The runtime does not expose a more specific model revision or context-window size. ## 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 (focused suites and Chromium regression; full local limitations 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>	2026-09-28 16:15:14 -07:00
3 people	0be2afcca6	feat(ui): improve task composer controls and pending input (#14322) ... ## Thinking Path > - Paperclip lets operators assign tasks to AI agents and review their work. > - The task composer controls the next message and its assigned agent. > - Operators needed a way to choose that agent's model and effort without leaving the composer. > - The old mode selector, upload button, and input cards made the mobile composer crowded and hid normal messaging during a pending decision. > - Harnesses publish different model and effort capabilities, so the picker must follow the selected agent. > - This pull request adds one responsive composer flow, keeps pending cards visible above it, and protects Codex ACP authentication in the local test path. > - Operators can choose run settings, send a message, and answer a pending card as separate actions. ## Linked Issues or Issue Description **Subsystem affected** Task composer UI, issue thread interactions, Codex ACP credential handling, and Storybook. **Problem or motivation** The composer did not expose model or effort for the selected agent. Mobile actions wrapped poorly. Pending questions and confirmations replaced the composer. A local Codex ACP test could also reuse host authentication after the managed key was removed. **Proposed solution** Put assignee search, model search, exact model IDs, effort, and fast mode in one picker. Use a mobile dialog. Replace the direct-upload plus action and separate mode selector with an Add menu and removable Plan or Ask chips. Place pending interaction cards above the usable composer. Keep these cards pending after an ordinary message unless their creator asks for comment superseding. Replace managed ACP auth files atomically and isolate the test key from host credentials. **Roadmap alignment** ROADMAP.md does not list an overlapping composer milestone. This change improves the existing task and review flows. ## What Changed - Added the combined assignee, model, and effort picker to both task composers. Search matches agent name, role, and harness. The server uses a curated Codex list by default and honors instance-declared models. Manual IDs remain available. - Added an effort slider for known model capabilities, a conditional Codex fast control, and reset. The picker opens in a modal on mobile. - Added the Add menu for files, supported goals, Plan mode, and Ask mode. Plan and Ask are exclusive removable chips. Keyboard mode cycling remains available. - Adjusted mobile spacing, avatars, wrapping, and Send placement. Removed the composer divider. - Moved pending question, confirmation, review, and related cards above the composer. Ordinary comments now leave question and confirmation cards pending by default. The onboarding prompt retains explicit comment superseding. - Updated the Storybook composer group with responsive states and the production picker. Added UI, service, route, and browser regression coverage. - Isolated Codex ACP API-key authentication, skipped subscription auth merge and shared-home copy-back for remote API-key runs, and replaced the managed auth file atomically. ## Verification - `pnpm -r typecheck` — passed on the final local head. - `pnpm check:token-gates` — passed on the final local head. - `pnpm exec vitest run server/src/__tests__/adapter-models.test.ts ui/src/components/task-chat/ComposerRunSettingsPicker.test.tsx` — 31 tests passed, including role and harness search, declared Codex models, and filtering general OpenAI models. - `pnpm exec vitest run server/src/__tests__/issue-thread-interactions-service.test.ts` — 74 tests passed. - `pnpm exec vitest run packages/adapters/codex-local/src/server/acp.test.ts` — 42 tests passed, including remote API-key copy-back isolation. - `pnpm test:run` — attempted locally; the embedded PostgreSQL test database could not initialize on macOS. The isolated `heartbeat-run-event-sequencing` suite reproduced that environment failure. GitHub CI runs the full test matrix for this head. - `pnpm build` — passed on the final head. `pnpm build-storybook` passed after the last UI change; only server code, tests, and docs changed afterward. - Live local test drive — Codex ACP ran a task with a managed API key. The test agent was restored to its default ACP configuration afterward. - Review the interactive stories under the top-level Composer group with `pnpm storybook`. Check a narrow desktop width and mobile Plan, Ask, picker, and pending-question states. ## Risks - A pending card stays open when an ordinary comment changes the discussion. Its creator can set `supersedeOnUserComment: true` when a new comment should replace it. - Model and effort overrides persist on the task until reset or changed. An unlisted manual model ID may fail when the provider runs it. - Some harness catalogs do not report effort support. The picker hides effort for those models. - No database migration is required. > 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 OpenAI GPT-6 via Codex. This runtime does not expose the exact model ID or context window to the task. The model used code execution and browser tools. ## 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 - [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> Co-authored-by: OpenAI Codex <codex@openai.com> canary/v2026.928.0-canary.10	2026-09-28 23:02:39 +00:00
DottaandPaperclip	24c58e479a	Improve task artifacts with rich cards and editable stories (#14469) ... Render eight artifact card types from real task records and share them with editable Storybook stories. Preserve document review and media/file actions, and load bounded CSV previews on request. Co-Authored-By: Paperclip <noreply@paperclip.ing> canary/v2026.928.0-canary.9	2026-09-28 15:38:06 -05:00
DottaandPaperclip	18e8c121d9	fix(runner): include Grok support in public installs with sandbox prerequisites (#14024) ... ## Thinking Path > - Paperclip manages agents through a shared native runner. > - Built-in harness support should ship with Paperclip's public distribution. > - Grok already speaks ACP; it does not require a new public bridge package. > - Sandbox provisioning owns the native executable and its pinned version. > - The runner must verify that prerequisite without downloading it during npm installation. > - This change separates built-in launcher identity from external runtime identity. > - Clean npm installation and live staging checks verify the distribution boundary. ## Linked Issues or Issue Description Refs #13882, #13973, #13977, #13979. This follow-up now targets master after #13882 was squash-merged. It replaces the private `@paperclipai/grok-acp` workspace package with runner-owned assets. Current master is included so the branch also contains the merged scheduler, complete-event capture, and durable cleanup fixes. ## What Changed - Ship Grok launcher and qualification metadata inside the runner's compiled output and the public server's vendored runner tree. - Remove the separate Grok npm package and all package-manager install hooks for this runtime. - Require the checksum-verified Grok Build 1.0.13 binary at `/opt/paperclip/providers/grok/1.0.13/grok` in the selected execution environment. Provision it explicitly in the Daytona image and CI setup. - Keep native binaries outside the provider pack. Bind the built-in launcher into the pack manifest. - Preserve executable leases, descriptor-backed startup, credential fences, permissions, and exact ACP model admission. - Use `builtin:grok-acp` and `native:grok` as profile identities. Historical package-profile sessions fail closed on resume rather than being silently reinterpreted. - Resolve built-in assets from the authenticated sidecar location, including public server npm layouts. Keep the controller path out of provider environments. - Add clean npm tarball installation verification to the existing trusted canary CI job and the admitted manual EC2 verification path. It stages a unified release version and runs npm lifecycle scripts, then verifies missing-prerequisite rejection and admission after separate provisioning without credentials or inference. - Include the controller-owned provider pack in stamped Cloud images. Unstamped local images omit the pack and remain usable; remote ACPX requires full source provenance. - Correct CLI approval-page metadata for an already authenticated Cloud board user; approval authorization remains unchanged. - Honor explicit native-runner enablement in the Cloud agent picker and direct setup page, keeping the flag disabled by default. - Allow selecting the execution environment before connecting credentials. Include Grok in the existing authenticated hello-probe flow, targeting its pinned native prerequisite for runner setup. - Recover an existing subscription sign-in conflict through an explicit cancel-and-retry action, serialized after cancellation succeeds. - Preserve the selected ACPX harness before normalizing config fields, so new Grok agents use the Grok default model. - Keep the credential-free Cloud provider pack root-owned and readable after runtime UID remapping; verify manifest and referenced asset access under an unrelated unprivileged UID during image builds. - Archive prior failover backups alongside explicitly replaced harness state, preserving evidence while preventing stale backups from blocking a fresh replacement. - Update Daytona image content inputs and contract tests for the built-in assets and explicit provisioner. - Document and regression-test the shared `approve-all` default for Grok setup, saved configuration, and native execution. Explicitly saved restrictions remain unchanged. ## Verification Current merge-repair head `df09eb3e1a619430ad8419a0ee9aedd486689b05` incorporates master `f1a394bd30cb56fb9e479f98b9f50176fe921858` after the base PR was squash-merged. All 12 conflicts came from incoming files identical to the tested pre-squash base. The final tree exactly matches a three-way merge using that original base, preserving built-in Grok distribution and removal of the obsolete private package. All 252 focused runner/UI tests, six npm-isolation tests, and token gates pass. Fresh exact-head Greptile review is 5/5 with no outstanding findings; security scans and EC2 native compilation pass. All current-head CI is green: 56 successful checks/statuses and four intentional skips ([run 36468768035](https://github.com/paperclipai/paperclip/actions/runs/36468768035)). The repository owner explicitly authorized bypassing code-owner approval after all checks passed; no CI checks or repository protection settings are bypassed or changed. The only remaining PR was removed from the completed stack metadata to permit native auto-merge. Earlier integration head `78cb306ecc41b5c96577c26c1d89153b0ef865a1` includes master `3447609d2247e75e55d91493dda91a608364f672` (2026-09-28). Two master advances during verification overlapped the eval catalog; the final merge preserves Grok qualification, completion updates, and bounded API-response reading in all 348 cells. All 77 focused catalog/eval/workflow tests pass. Both native stack layers (#14397) are mergeable, and both exact-head Greptile reviews are 5/5 with successful security scans and no unresolved review threads. All current-head CI is green: 56 successful checks/statuses and four intentional skips ([CI attempts](https://github.com/paperclipai/paperclip/actions/runs/36447124691)). The initial attempt lost two EC2 runners to shutdown signals and stalled a third shard during dependency preparation; all three passed the same-commit failed-job-only retry. Trunk code-owner requirements remain enforced. The review summary’s non-blocking saved-asset offset classification note concerns code already merged in #14301; those runtime files are identical to master and outside this stack’s diff. Historical live evidence below retains its original source revisions. [Final public npm verification](https://github.com/paperclipai/paperclip/actions/runs/36445542764) passed on `76ea70cd4d13786a042af9df82f0fd7a8c85ae30`: 17 public packages, an executed offline lifecycle sentinel, unchanged consumer lock, built-in launcher, missing-prerequisite rejection, and verified separately provisioned binary/command lease. Provisioning and cleanup require no host privilege elevation; only the positive probe mounts the temporary native binary read-only. The verifier is unchanged by the final master merge. All six isolation tests and an offline npm smoke test pass. The prior head had 56 green CI checks and a 5/5 review after two unchanged tests timed out and passed a failed-job-only retry ([CI attempts](https://github.com/paperclipai/paperclip/actions/runs/36444597313)). All 56 recovery-display/lineage tests pass; re-review cleared the already-covered missed-retry concern. Earlier EC2 failures remain retained: [npm lockfile rejection](https://github.com/paperclipai/paperclip/actions/runs/36436311203), [missing compiler in the slim image](https://github.com/paperclipai/paperclip/actions/runs/36440210984), and the aggregate 15-minute test timeouts in those broad runs. Both broad attempts passed typecheck, token gates, Product E2E type/unit checks and build. The focused EC2 lane preserves the existing trusted-actor and immutable-source gates. Earlier documentation/test checkpoint `ff244c4fd78a7ede5a3e00efe09f475f133ef33e` leaves runtime behavior unchanged. 154 focused tests pass across configuration building, native provider resolution, permission policy, credentials, UI configuration, and new-agent setup (including both Grok auth modes); token gates pass. All fresh CI is green for this head: 56 successful checks/statuses and two intentional skips ([run 36367065119](https://github.com/paperclipai/paperclip/actions/runs/36367065119)). Greptile is 5/5 with no new findings. Grok already inherits the shared `approve-all` default, so unattended setup requires no manual permission change. Runtime head `bb5a9307991f1ac567b781970ef11b39d518e19b` fixes a final staging continuation failure before provider startup: explicit replacement archived the old harness but left its failover backups active, which caused `runner_harness_state_mismatch`. The regression fails before the fix and passes after it; all eight adjacent recovery-safety cases also pass. Old backups remain inspectable inside the continuity archive. All fresh CI is green at this head ([run 36360839248](https://github.com/paperclipai/paperclip/actions/runs/36360839248)), with a 5/5 review. One unrelated Cursor test timed out in the initial server shard; the same-commit failed-job rerun passed, and both attempts are retained. Staging deployment is confirmed healthy on this revision. The controller image is `ghcr.io/paperclipai/paperclip@sha256:6ad91c487910ccd2596ff7aed0a3a3ea5233d12b51b83cd6e1402237749b9673`. The final browser-created staging task passed on this exact revision with API authentication: context read → structured human question → controller restart → answer submission → same native provider session resumed → document saved → task Done. The two turns took approximately 119s and 77s. The actual write receipt was applied, and the saved document has exactly one revision containing the selected answer and requested marker. Usage and cost were not reported. [Controller image build](https://github.com/paperclipai/paperclip/actions/runs/36360889243). - Previous integration head `a44f7dbb6b6f77cd9ed893756ca453307f281e5f`: all CI green (53 successful checks/statuses, two intentional skips), including repository typecheck/build/tests, native Runner tests, browser shards, and canary installation checks. [CI run 36358672529](https://github.com/paperclipai/paperclip/actions/runs/36358672529). Greptile is 5/5 with no unresolved findings. - Focused checks cover Grok credentials, executable admission, launcher assets, provider-pack paths/permissions, workflow contracts, setup defaults, CLI authorization, and subscription conflict recovery. All 39 protocol definitions validate. Final integration checks pass 124 catalog/evidence/cache tests and nine project-form tests; token gates pass. Some local dependency checks could not load the stale installed dependency tree; the corresponding fresh EC2 checks pass. - Clean public npm installation passed on EC2 at `8b172ebcf8e02e30662d830c00f3961e3bd459ec` ([run 36164964900](https://github.com/paperclipai/paperclip/actions/runs/36164964900)): 17 unified-version packages, lifecycle scripts enabled, built-in launcher present, no separate Grok package or npm-downloaded binary, missing prerequisite rejected, separately provisioned native executable and command lease verified. No credentials or inference were used. Subsequent changes preserve this npm asset layout. - The immutable Daytona prerequisite image is `ghcr.io/paperclipai/paperclip-daytona-runner@sha256:98957d5be0ac774d086b6402b5849e8e6356fec70fb8c09fca6eb4ed6de918e0`, built from `5a2db471f3ddabe77f9f80e76ed27f996cb97fba`. The previous Cloud controller image was `ghcr.io/paperclipai/paperclip@sha256:fd914e1ab1e45f741e8e078ff452d16f082d7ac05f9b4b3506d3a3c64150d204`, built from `a44f7dbb6b6f77cd9ed893756ca453307f281e5f`; it is superseded by the latest image above. Its EC2 build verified provider-pack access under an unrelated unprivileged UID. - Browser staging at `40f898bc4cba73c1dff4e6344a3983ba0fb247ef` passed full Grok onboarding with the correct `grok-4.7` model, saved credential delivery, and pinned Daytona execution. A browser-created task read context and asked the structured human question. After a controller restart, answering the persisted question resumed the same native provider session, saved the requested document, and completed the task. Actual tool outcomes and durable state agree: one question and one document revision. The two successful turns took 42.7s and 63.1s; usage and cost were not reported. - Restricted policy returned the expected `approval_required` outcome. Functional staging tests explicitly selected `approve-all`; controller authorization and governed approvals remain enforced. Temporary board CLI access was revoked and verified rejected (HTTP 401), and the disposable onboarding agent was paused. Failures remain retained: the pre-fix continuation failure (its task remains blocked; the passing final task is fresh), the original Cloud provider-pack permission failure, the expected restricted-policy denial, the superseded npm staging failure, and an earlier monolithic CI infrastructure timeout. Browser CI exposed a project alias/form race; the final stack uses master's stronger draft-preservation fix and all browser shards pass. Historical full subscription/API protocol and Product rosters retain their original source revisions and do not qualify this packaging revision. No local Docker or Rust build was used. ## Risks The branch includes master’s draft-preservation fix for project URL aliases. It keeps the same project’s edit form mounted and clears prior data when the project or company changes. Custom sandboxes and local execution hosts must provision the pinned binary before Grok starts. Missing, changed, unsupported-platform, and symlinked executables fail admission. The new builtin profile cannot resume sessions created with the former private-package profile. Existing Claude/Codex npm bridge profiles retain their package pins. Grok restricted modes preserve the selected policy but cannot automatically admit Paperclip calls: ACP permission metadata does not independently bind tool authority, so those calls stop with `approval_required`. New Grok configurations default to `approve-all`, including API configurations that omit the mode. Existing explicitly restricted configurations remain restricted; controller authorization and governed approvals remain enforced. ## Model Used OpenAI GPT-6 through Codex, with tool use and code execution. The exact serving 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 - [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> canary/v2026.928.0-canary.8	2026-09-28 14:54:44 -05:00
DottaandPaperclip	992f720262	fix: make runner task context ownership explicit (#13753) ... <!-- Write all pull request text in Simplified Technical English (ASD-STE100). --> ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Task descriptions, comments, continuation data, skills, and execution rules enter several agent adapters. > - The same source can be rendered by more than one automatic input carrier. > - Failed resumes can also rebuild input from stale or compact context. > - This pull request gives each Paperclip-owned source one delivery owner and preserves the required transport boundaries. > - It adds deterministic adapter, interaction, runner, and browser tests for these boundaries. > - The benefit is more predictable context delivery with explicit evidence for later live qualification. ## Linked Issues or Issue Description Related: #13144 removes a duplicate environment payload and bounds wake lists. Related: #11360 addresses Hermes resume behavior. This pull request preserves compatible active-session formats while repairing context ownership and stale question creation. **What happened?** Task descriptions and comments could enter more than one automatic context block. Native transports could wrap a complete model input in a second task envelope. Some legacy and gateway adapters could omit the owned assignment on ordinary tasks or rebuild a failed resume with stale compact context. A continuation could also request a question after newer human comments had arrived. **Expected behavior** Each task or comment source has one automatic model-facing owner. Distinct comment IDs and repeated wording remain distinct. Fresh fallback attempts rebuild the required full context. A question request is rejected when newer queued human direction makes it stale. Harness access policy remains owned by execution configuration. **Steps to reproduce** 1. Build a task with a description and current comments. 2. Capture the actual adapter or runner input. 3. Compare source ownership and task-envelope nesting. 4. Queue a human comment before a continuation requests a question. 5. Trigger a failed resume and inspect the fresh retry input. 6. Run the focused adapter, interaction, runner, and browser checks. ## What Changed - Add shared prompt-section selection at the provider-attempt boundary. - Deliver owned assignment context through native, legacy CLI, ACP, gateway, cloud, Pi, Kimi, Grok, Gemini, OpenCode, Cursor, OpenClaw, and Hermes paths. - Rebuild full or compact context after resume recovery changes the attempt. Add native and Claude ACP tests of actual recovery requests. - Preserve custom templates, loaded instruction files, execution policies, and older active-session formats. - Record continuation source metadata and reject stale question creation under the issue-row lock. - Add explicit Product E2E context-integrity profiles, prerequisite gates, credential-isolation checks, and report fixtures. - Bypass service-worker forwarding for same-origin Vite development modules. A real Chromium test fails with resource exhaustion before the repair and passes after it. Production asset caching keeps its existing policy. - Add browser diagnostics and service-worker module-loading regressions. - Add an explicit zero-retry eval option. The default retry behavior remains unchanged. Each campaign records its effective policy. - Remove the model-facing working-directory sentence from four prompt builders. Existing workspace, sandbox, permission, and custom-template configuration remains unchanged. - Align the everyday workflow assertion with the current 47-entry catalog. Compared with current upstream master, the branch carries the context-ownership implementation and its tests, the explicit context-integrity catalog and evidence harness, and the focused browser regression checks. ## Verification **Merge assessment:** focused regression evidence supports merge. This is not full completion of the original broad qualification matrix. The maintainer has authorized merge after fresh verification of the master integration. - Current head: `bbd52f82114eabf09bc7b1a7e97d54a5b43bbc00`. This integrates current master `2f585ef26a1814fa209715242d1ca791b63e4c4e`. All 14 conflicts are resolved. Cancellation checks, workspace finalization, native Grok support, and both sets of tests are retained. - Current-head Greptile: **5/5**, with no blocking findings. The review names this exact commit. All **59 reported checks are terminal: 55 successful, 4 skipped, zero pending or failing**. This includes the full root general and serialized suites, separate runner checks, typecheck, build, canary, browser E2E, Docker, and security checks. The successful legacy security status is included in that total. - After integration: workspace typecheck and full build passed. Separate runner checks passed: **2,160 TypeScript tests (10 skipped), 582 Rust tests, and 39 preparation checks**. Other passing checks include 621 Product E2E harness units, 376 focused shared/adapter tests, 160 real-database/API tests, 86 Hermes tests, 18 browser-support checks, and Product E2E typechecking. The complete root suite passed in CI. The duplicate local monolithic root run was stopped after that CI result; it is not counted as a completed local pass. - New native recovery coverage retains full assignment, completion contract, and explicit skill selection after safe replacement, for old and prepared input formats. Full native session test file: **136/136 passed**. - New Claude ACP coverage captures actual fresh, resumed, and missing-session fallback requests. It verifies one assignment copy, comment order, identical text under distinct comment IDs, and full fallback context. Full file: **33/33 passed**. Both affected TypeScript checks passed. - Existing deterministic tests cover source revisions, approval and trust boundaries, completion validation, custom templates, compatible sessions, standalone driver wrapping, and maintained adapter transport requests. - Provider-free browser support: **17/17 passed** after the master merge. Service-worker unit tests: **33/33 passed**. The module-overload regression failed before the repair and passed after it in real Chromium. ### Fresh live comparisons The new batch ran exactly four Product E2E attempts. **All four passed on the first attempt; no retries.** Each has six terminal matchers plus the existing browser lifecycle and invariant checks. | Exact case ID | Control | Candidate | |---|---|---| | `core-compatibility.runner-codex.local.plan-revise-accept` | Passed | Passed | | `local-session-integrity.runner-acpx-claude.local.structured-question-restart-resume` | Passed | Passed | The plan case checks a revised canonical plan and revision-bound approval before completion. The question case restarts the server before submitting the answer, then verifies the continuation completes. Control source is `dfa4e1bda8d50a1a01746603251a9128dbe9d0d6`. Candidate source is `79fcdb5dece501d28064ea9da306603881b46f0c`. They use identical frozen definitions and provider versions: Codex `0.156.0` with `gpt-5.6-sol`; ACPX `0.13.1` / Claude ACP `0.73.0` with `claude-sonnet-5`. The September 24 head added master browser recovery and test-only changes. The September 28 head also integrates newer master changes, including cancellation, workspace finalization, and native Grok. These are frozen-source live results, not exact-head live runs. The candidate received one description copy where the control initially received three. The submitted initial plan envelopes were 7,969 versus 19,097 characters. Question envelopes were 7,592 versus 18,919. These are structural measurements, not whole-provider token or dollar savings. ### Earlier evidence and failed attempts - The preceding fresh batch has four effective passing pairs: OpenCode comment continuation and assigned skill, native Codex comment continuation, and native Claude comment continuation. It retains **11 attempts: eight passed and three failed**. - Original failures remain recorded: missing local PostgreSQL library links before task creation; host-sleep cleanup after task/page checks passed; and a Claude **control** session-open rejection before a model turn. Setup was repaired identically on both worktrees. The permitted unchanged infrastructure retries passed. The underlying Claude provider startup error was not retained and remains unknown. - Older R2 retains **17 passes and one failure** across 18 attempts, including eight both-pass native/legacy Codex/Claude pairs. Its OpenCode blank-page failure led to the service-worker repair. R2 is historical evidence: master changed the native fixed prompt and removed duplicate wake environment data afterward. - The September 24 CI run initially failed one unrelated preview readiness test (`ECONNREFUSED` on its local fixture). Its test and production code match master. Isolated local verification passed **28 tests, 3 skipped**. One unchanged CI retry passed the full shard: **831 passed, 1 skipped**, including all **31 preview-exposure tests**. The aggregate CI gate passed afterward. The precise startup cause remains unknown; a port race is a hypothesis, not a proved cause. ### Limits The original wider profile/workflow matrix, repeated trials, and remote Daytona qualification are incomplete. These results support a focused merge recommendation, not statistical equivalence or universal harness qualification. Some usage receipts are missing in both variants, so no token or dollar savings are claimed. The $500 ceiling was preserved using conservative allowances; failed attempts and unknown charges remain in the ledger. Reproduce the focused additions with `pnpm exec vitest run packages/adapters/claude-local/src/server/acp.test.ts` and `pnpm --filter @paperclipai/paperclip-runner exec vitest run src/native-session-runtime.test.ts`. Full checks use `pnpm -r typecheck`, `pnpm test:run`, `pnpm build`, and the separate runner checks. Paid evals require the frozen definitions, profiles, and credentials; do not use `--all` as a substitute for the selected cases. ## Risks - Context placement changes can affect model behavior. Deterministic checks cover the selected paths, but live qualification remains incomplete. - The stale-question guard can reject a request when queued human comments arrived during the run. This is intended. - New stored inputs and model envelopes retain compatibility readers for older active sessions. - Custom templates may intentionally repeat content. - Removing a model-facing working-directory sentence does not change filesystem, command, sandbox, or permission configuration. - The worker bypass applies only to same-origin development module paths. Cache-policy tests preserve private-response handling and production asset caching. Mounted HTTP fixture changes remain test-only. - This PR does not claim measured token savings or statistical equivalence across every harness. ## Model Used OpenAI Codex, exact model gpt-6-astra, with repository tools and code execution. Bounded supporting work used gpt-5.6-luna and gpt-6-luna. The serving context-window size is 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 described the issue in-PR using the required issue fields - [x] I have not referenced internal/instance-local Paperclip issues or links - [x] My branch name describes the change and contains no internal ticket id - [x] I have run the focused local checks and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect these changes - [x] I have considered and documented risks above - [x] All current-head Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups for the current head - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> canary/v2026.928.0-canary.7	2026-09-28 14:49:14 -05:00
DottaandPaperclip	2f585ef26a	fix(ui): preserve newer drafts after repeated receipt cleanup (#14332) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The task composer keeps unsent text across page reloads. > - A server receipt confirms a submitted message after a reload. > - Replayed cleanup can apply the old text offset to a newer draft twice. > - This pull request reconciles each confirmed attempt once and preserves both tabs’ unsent intent. > - The full newer draft stays available for the next send. ## Linked Issues or Issue Description **What happened?** Reloading the classic composer while a save was pending could truncate a newer draft. Replayed receipt cleanup changed `A newer draft written while delivery was pending.` into `nding.`. The storage helper rejected the duplicate settlement, but the effect still changed the editor and its body reference. **Expected behavior** A receipt settles its retained submission once. Later cleanup must preserve the newer draft in both the editor and browser storage. **Steps to reproduce** 1. Send a comment and hold its HTTP response after the server accepts it. 2. Type a newer draft, then reload the page. 3. Restore the matching receipt under React StrictMode. 4. Inspect the newer draft after effect replay and unmount. The deterministic regression reproduces this on current master. The existing browser test exposed the issue during #14329 verification. Related draft persistence code came from #13338. A search found no separate fix for this duplicate settlement. ## What Changed - Reconcile each confirmed attempt once in memory so duplicate effects cannot trim its newer draft again. - Unlock confirmed drafts when storage writes fail. Never acquire another tab’s pending receipt or assume its text is newer. Keep a conflicting local draft and its attachments in tab-scoped session storage while preserving the shared draft unchanged. - Add component regressions for StrictMode replay, exact editor and stored bytes, foreign receipts, attachment-only differences, reload recovery, unavailable storage, and task navigation before autosave. A brief notice explains when this tab has a separate draft. ## Verification - RED: the new test received `nding.` instead of the full newer draft before the fix. - Review RED: three cases reproduced a locked composer after failed storage writes or another tab's settlement. A further negative case preserves a newer retained attempt and its attachments. - Cross-tab review RED: four deterministic cases reproduced lost local text, foreign receipt takeover, attachment loss when text matched, and overwriting newer stored text. - GREEN: 118 tests across `IssueChatThread`, `composer-draft`, and `comment-submit-draft`, including same-mounted A → B → A navigation and a full recovery-storage failure. Two further lifecycle RED tests verify finishing a recovery returns to the shared draft on a later visit while continued typing and attachments retain recovery. - Both existing browser reload cases passed against a fresh server and database on exact head `ceb80aca77fc8cc0f813c328ba87025b3e1a2222` (42.5 seconds), with classic mode enabled and disabled. The browser flow verifies one original comment, a preserved draft, and a successful second send. - UI typecheck, token gates, and the shipped static UI build passed. The initial typecheck required the fresh worktree's plugin SDK build; the retry passed after that dependency built. - The session-only failure mock also preserves localStorage on platforms where both share the Storage prototype; this fixes the Linux workspace test failure. - All 56 checks passed on `4c30dcccc4fe5b90dcd08dd1d90475ea89e4a376`; Greptile is 5/5 with zero open review threads. An unrelated runner baseline scan hit its existing 100 ms deadline once; its isolated test and the single failed CI job passed on retry without source changes. This four-file UI fix does not change server or database code. - Final alternate-staging acceptance passed on deployed source `d884e1ab046cc76004e35e6091e9e6e2c918c9eb`, with exact health checked before and after. Three real browser cases used shipped static UI and real API saves: reload during an accepted-but-unacknowledged save in both composer variants, plus two classic tabs with different drafts. The test deliberately held only its own accepted POST acknowledgement and temporarily withheld its own GET receipt from one tab to reproduce settlement ordering. Both drafts survived independent reloads, the shared draft remained intact after the other tab sent, and finishing recovery returned to the shared draft on a later visit. Exact request IDs/counts and server comment bytes passed; no provider was invoked. Screenshots preserve the live drafts before fixture cleanup. - An ordinary authenticated Chrome/CUA visit independently showed the exact original and newer comments; an additional unsent draft survived reload and was then cleared. All three isolated fixtures are complete, browser contexts are closed, and original user tasks were untouched. ## Risks - Reconciliation uses the exact request ID and draft key. Conflicting drafts stay separate; the tab recovery survives same-tab reload through session storage and does not outlive the browser tab. If recovery storage is unavailable, the editor keeps its in-memory text and explicitly warns the user to copy it before leaving. - Attachment selections remain bounded to 20 per draft; if combining equal-text snapshots would exceed that limit, each original selection stays in its respective draft. - No API, schema, or migration changes. The status notice uses existing design tokens. ## Model Used OpenAI Codex, GPT-6, with tool use and code execution. The runtime does not expose the exact deployment variant or context-window size. ## 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>	2026-09-28 14:03:36 -05:00