758 Commits

Include renames

Author	SHA1	Message	Date
Devin FoleyandPaperclip	1d5a439e2b	Diagnose native model rejections and preserve repairable reviews (#15304) ... ## Thinking Path > - Paperclip manages agent execution and recovery. > - A native provider can return a failed turn with an accepted result. > - This path replaced a structured model rejection with a generic failure. > - Finalization could also queue the same failed request for recovery. > - This change retains a bounded model and authentication diagnosis. > - Workers wait for a configuration fix; reviews retain their assignment. ## Linked Issues or Issue Description **Describe the bug** A native Codex turn can report HTTP 400 because the selected model is not supported by its ChatGPT connection. When the turn also returns a failed semantic result, the run shows `Native session failed` and `adapter_failed`. The native failed-result finalizer can create a retry intent before the adapter error is saved. **To Reproduce** Use a native Codex session that returns a structured `invalid_request_error` model/ChatGPT rejection and an accepted failed result. The regression tests reproduce this with a stub provider and a real isolated PostgreSQL database. No provider account or live request is needed. **Expected behavior** Show an actionable error and stop automatic retries of the rejected request. Block the current worker for a configuration fix. Preserve a pending review so its reviewer can explicitly retry after repair and resolve the original assignment. A stale run must not block a new task owner or override a completed review. **Version and deployment mode** The affected source path is present at `858094ba81`. This change applies to native Codex execution in local and remote environments. Related work: #13853 covers CLI protocol compatibility, #14942 refreshes runtime/model pins, and #12767 proposes legacy adapter model validation. This change concerns accepted native failed results and does not replace those changes. ## What Changed - Recognize only a bounded structured HTTP 400 provider error for the exact selected model and failed turn. - Save a fixed actionable error and four fixed diagnostic fields. Exclude raw provider text and model names from the new diagnostic. - Re-derive finalization evidence from the pinned execution and committed runner event, including its canonical hash and identity bindings. - Block only the current worker. For a current pending reviewer, preserve the original review decision and status version, release execution, and create no automatic recovery action. Recheck execution ownership under the issue lock. Preserve superseded and completed reviews. - Keep unknown failures, accepted results, cleanup, and ownership fences intact. Advance the status policy version to `phase6-v8`. - Route this exact failure code through the existing configuration recovery path. Permit an authenticated Board retry of this native failure only for its saved, still-pending review; discard caller review markers and revalidate at dispatch. ## Verification - Node 24: `pnpm -r typecheck` and `pnpm build` passed. - Independent focused validation: 222 tests passed across the new observer and PostgreSQL integration suites, status arbiter, diagnostic redaction, recovery classification, wake policy, and wake use cases. - The integration tests use actual result persistence, finalization, and the wake dispatcher. They cover zero retry dispatch for an authorized rejection, repeated reconciliation, pending/superseded/resolved reviews, reassignment, and a same-agent successor at an unchanged status version. - Final-commit verification: full typecheck and build passed; independent route/integration/arbiter/review validation passed 136 tests with no remaining source findings. This includes authenticated retry refusals, actual heartbeat retry admission, atomic queued-review claim, and acceptance of the original review. - The local aggregate test run was stopped to avoid duplicating the canonical Linux CI aggregate on this development host. Its partial log showed no failures; no full local-suite pass is claimed. The full hosted Linux CI aggregate passed on the final commit. - No live provider calls, runtime changes, or schema changes were made. ## Risks The classifier is intentionally narrow. A changed provider response format remains an unknown failure. No global model restriction is introduced. The worker blocker requires current task authority and does not replace newer assignments or completed review decisions. A pending review remains in review and requires an explicit retry after configuration repair; it is not rebound to a new status decision. The retry decision is durable. There remains a small existing gap between status/effect commit and diagnostic metadata persistence; a crash there can leave a generic diagnostic while preserving the blocked decision. No schema migration is required. ## Model Used OpenAI Codex, GPT-6, with reasoning, tool use, and code execution. The exact serving model suffix 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 #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 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 - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 20:57:33 -07:00
Devin FoleyandPaperclip	d21fe31243	Retry pooled execution projection reads after disconnects (#15300) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Task run lists show current execution and recovery state. > - Their execution projection reads runs, finalizations, pending interactions, and recovery records. > - A temporary database disconnect in one lookup makes the run list fail. > - This pull request uses the existing bounded retry policy for these read-only lookups. > - It keeps transaction reads, authorization, and liveness writes outside the retry scope. ## Linked Issues or Issue Description Refs #15248. This extends the same bounded SELECT retry policy to the execution projection used by task run lists. A wrapped `CONNECTION_CLOSED` error from the pending-interaction SELECT makes `GET /issues/:id/runs` fail. These lookups are safe to repeat on the pooled database connection. The shared projection helper also serves transaction callers, so retries must remain an explicit opt-in. ## What Changed - Opt the activity run list into per-query projection retries. Keep the default path single-attempt for transaction callers. - Rebuild only the failed SELECT. Use the existing three-attempt limit and 50/100 ms delays. - Keep completed reads, route authorization, and liveness backfill writes outside the retry callbacks. - Test each query, exhaustion, exact error preservation, non-transient errors, empty results, transaction defaults, run-list backfill behavior, and denied route access. - Document the pooled-read boundary. ## Verification - Frozen install passed with Node 24.21.0 and pnpm 9.15.4. - 92 focused tests passed across execution projection, activity service/routes, prior service-read retries, and the shared retry policy. - Independent review passed 56 projection, retry-policy, and route authorization tests with no blocking findings. - `git diff --check` and the staged diff and PR-text secret/PII scan passed. - Full workspace `pnpm -r typecheck` and `pnpm build` passed. - Greptile scored exact head `78783e84a5` 5/5 with no findings or open review threads. - Initial CI had one Runner test timeout: `stops and closes a timed-out consumer even when the caller requested a warm session` exceeded 5 seconds. Its fixture uses a 1 ms session timeout. The Runner package is unchanged from the base. The focused case passed locally (1 passed, 138 skipped). One unchanged-head rerun of that job passed all 1,219 Runner tests and the additional native-runtime integration test. All other CI lanes passed on the initial run. - Full Linux CI passed on exact head `78783e84a5`: 53 successful checks, two intentional Storybook skips, and no pending checks. It provides aggregate test coverage. A full local aggregate was not launched. The same host's prior aggregate exposed three known macOS immutable-cache rename failures in unchanged company-skills code, documented in #15298. ## Risks - An affected lookup can take up to two additional connection attempts plus 150 ms of delay. Persistent failures still return the final error. - The opt-in is only used by the pooled activity read path. No transaction, mutation, full request, or provider action is replayed. - No schema or response-shape changes. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository editing, code execution, and independent agent review. The exact deployment model ID and context window 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 references) - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass (focused suites; aggregate coverage is described above) - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 20:09:18 -07:00
Devin FoleyandPaperclip	858094ba81	Fix browser polling for unsaved agent chats (#15298) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agent Chat uses the shared task surface and browser panel. > - An unsaved chat has a temporary ID instead of a stored task UUID. > - The browser poll sent that ID to a PostgreSQL UUID query and returned a server error. > - This pull request waits for a saved task and validates task IDs before the query. > - Saved chats keep their browser tools and access checks. ## Linked Issues or Issue Description Refs #14331. That report describes the same draft-ID problem in email requests. This change covers the browser routes; it does not change email behavior. Opening an unsaved Agent Chat sends `chat:<agent-id>` to `/issues/:issueId/browsers` every three seconds. PostgreSQL rejects this as a task UUID. The draft is only a UI view model. The first send or upload creates its stored task. ## What Changed - Disable browser polling for unsaved chat IDs. Resume polling when the chat has a task UUID. - Return the existing task-not-found response for malformed task IDs before database access. Keep board authentication first and company and credential checks unchanged. - Test all six task browser routes, draft-to-saved polling, normal tasks, saved chats, and denied access. Keep canonical UUID v4, v7, and nil values queryable. - Document the draft and saved-chat browser contract. ## Verification - `pnpm install --frozen-lockfile` passed with pnpm 9.15.4 and Node 24.21.0. - `pnpm exec vitest run server/src/__tests__/browser-use-route-scope.test.ts server/src/__tests__/browser-use-connection.test.ts` passed all 35 tests. - `pnpm exec vitest run ui/src/hooks/useTaskBrowsers.test.ts` passed all 5 tests. - Independent review passed 14 hook and route tests plus both real task and saved-chat authorization cases. - `pnpm check:token-gates` and `git diff --check` passed. - Full workspace `pnpm -r typecheck` and `pnpm build` passed. - Full Linux CI passed on `dafff44a32`: 53 successful checks and two intentional Storybook skips. This includes all general and serialized test groups, UI and browser tests, typecheck, build, and the canary dry run. No CI retries were needed. - The full local `TMPDIR=/private/tmp pnpm test:run` started but did not complete. It was stopped after full Linux CI passed. Before the stop, three company-skills cache cases failed on macOS. A focused rerun reproduced all three as `EACCES` while renaming immutable cache staging directories. The test, company-skills service, and runtime cache files are byte-identical to the baseline previously reproduced on `e99854249c` for #15291. Other local groups were not reached; the full Linux CI run provides aggregate coverage. - Greptile scored the exact head `dafff44a32` 5/5 with no findings or unresolved review threads. ## Risks - Malformed task IDs now return 404 instead of reaching PostgreSQL and returning 500. The route does not map draft IDs to agents or tasks. - No schema, browser provider, task creation, or authorization policy changes. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository editing, code execution, and independent agent review. The exact deployment model ID and context window 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 references) - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass (focused suites; full local run limitations are reported above) - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 19:41:52 -07:00
DottaandPaperclip	45862dd210	docs: add a product feature map (#15288) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Its user journeys span tasks, agents, projects, connected apps, governance, and CLI operations. > - Existing tests do not provide a shared index of entry points and verification steps. > - Contributors need to see which surfaces a change affects and what evidence exists. > - This pull request adds an optional product feature map with recipes and explicit coverage gaps. > - The map adds no CI checks or required maintenance for future pull requests. > - Contributors can use it to find verification steps and state what they checked. ## Linked Issues or Issue Description **Issue type** Missing documentation. **Where is the issue?** User-journey verification guidance in AGENTS.md and doc/DEVELOPING.md. **What's wrong?** There is no shared index of product features, user entry points, available test evidence, and remaining coverage gaps. Shared components can hide differences between their hosts. **Suggested fix** Add a documentation-only capability index and verification recipes. The format takes inspiration from [Omnigent's feature map](https://github.com/omnigent-ai/omnigent/tree/91acfbbb59f6fc210ff95a9e9428aadd62e06582/feature-map). The recipes describe Paperclip's own behavior and tests. Searched GitHub PRs and issues for `feature map` and `feature-map`. No duplicate change was found. Checked ROADMAP.md. This PR documents existing capabilities. ## What Changed - Added 35 feature recipes, 171 named sub-features, and 91 entry points across product, CLI, operator, and developer surfaces. - Each recipe describes setup, expected results, existing automated evidence, manual verification, and coverage gaps. - Added a dated source snapshot of 185 non-test page TSX modules in 15 areas. The snapshot describes documentation coverage, not runtime health. - Included entry points within existing pages, CLI/API operations, and experimental surfaces. Identified helper-only test evidence where a rendered journey has no automated proof. - Linked the map from AGENTS.md and doc/DEVELOPING.md as an optional reference. - The final diff contains only Markdown and the inventory JSON. It adds no checker, tests, package commands, workflows, dependencies, scheduled work, or mandatory inventory updates. ## Verification - PASS: local documentation links resolve and the inventory JSON parses. Confirmed the final PR diff contains only 39 documentation files. - PASS: `node --test '.github/scripts/tests/*.test.mjs'` — 381 existing tests after removal of the feature-map tests. - PASS: `git diff --check`. - Earlier local build and typecheck passed. The full local test run was stopped after 26 minutes with failures in unchanged chat-channel and native-runner integration tests. It did not complete. These application checks were not repeated for the documentation-only removal. - [CI on the preceding head](https://github.com/paperclipai/paperclip/actions/runs/37398407523) passed all applicable checks. Checks on the final documentation-only head are pending. - PASS: Greptile review on final head `45c4b0aeb26540324825da43e81bee2336ad1c1f` is 5/5. There are no unresolved findings. - Live product/provider journeys were not run to author the map. The recipes identify available evidence and manual steps, not new qualification results. ## Risks Low product risk: the PR changes documentation only. Recipes and the source snapshot can become stale. Maintenance is optional and based on review. Linked tests do not prove that every documented journey works. The map states remaining coverage gaps. ## Model Used OpenAI Codex, GPT-6 family, with repository inspection, reasoning, tool use, and code execution. The exact serving model ID and context window 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 - [ ] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-06 02:08:47 +00:00
DottaandPaperclip	0e0b63e5a5	feat(connections): add experimental task-pinned AI routing (#14967) ... ## Thinking Path > - Paperclip manages AI agents and their work. > - AI Connections separate account access from models and harnesses. > - A pool must act as one connection while retaining each task’s account. > - Core must enforce member access and preserve session and recovery rules. > - A plugin supplies rotation policy without receiving credentials. > - This change adds durable routing and native connector setup and management. ## Linked Issues or Issue Description **Subsystem affected** AI Connections, Connectors, plugins, run dispatch, and session compatibility. **Problem or motivation** Operators need to rotate new tasks across saved accounts while each task keeps its account and session. Pool setup must fit the existing connector catalog and account workflow. **Proposed solution** Add an experimental router binding, a capability-gated plugin hook, and transactional task pins. Plugins declare native pooled connectors through `aiConnectionRouter`. Core hosts the existing-account picker, ordering step, and account settings. Related usage contract: #14936. Companion private plugin: https://github.com/paperclipai/paperclip-cloud/pull/643. **Roadmap alignment** This extends Apps and AI Connections. Core supplies generic enforcement and native connector UI; the private plugin owns rotation and quota policy. The prior duplicate search found no matching router implementation. ## What Changed - Add a router binding without changing existing concrete bindings. Keep the instance flag and new pools disabled by default. Require manual operator configuration. Show no routing toggle in Experimental settings on either open-source or Cloud installs, even after routing is enabled. - Persist company-scoped pools, one shared cursor per pool, and pins keyed by company, pool, agent, and task. Commit pins and cursor advances together with revision checks and bounded retries. Persist run-ID affinity before allocation. - Pass only authorized metadata and normalized usage to plugins. Core retains credential handling, member access checks, runtime qualification, and recovery evidence. Probe outside locks with a shared 15-second budget and freshness cache. - Resolve routing before credential preparation and backend selection. Preserve pins through turns, session resets, removed members, and quota waits. Retain admitted recovery after disable or uninstall. - Separate credential session epochs from token generations. Verified refresh preserves the epoch; reconnect and manual replacement change it. Include the credential slot ID in session and usage-cache identity, so reconnecting an indexed legacy account invalidates its old session even when both epochs are zero. - Validate pool member installations before accepting saved-agent bindings and recheck compatibility when the harness changes. Install only authorized members in the new-agent transaction and record their IDs in local activity. Pool membership cannot install a restricted shared connection. - Preserve pool bindings when agents hire teammates through either creation API or native caller runtime inheritance. Block stale manager credential references; retain explicit child authentication precedence and reject incompatible inherited pools. - Add native connector registration through plugin metadata. Reuse the Connectors catalog, setup header, account header, sidebar, dialogs, and usage display. Setup selects and orders saved connections. Advanced settings hold usage rules and member runtime defaults. New-account setup opens in another tab. - Use revision-checked pool archival from the Connectors catalog and account page. Keep task pins, cursors, recovery evidence, and underlying connections. Reject ordinary connection updates or removals that bypass pool revisions. - Add pool selectors, composer models, override notes, quota status, run details, activity records, and local run-log records. Keep session-adoption copy minimal. - Show **Used by** below the pool connections. List current company agents with shared avatars and profile links. Include paused agents; exclude terminated agents and agents using another pool. - Add Core stories for the generic connector workflow and runtime surfaces. Cloud stories reuse these production routes and tokens through a preview-only alias. ## Verification - Final head `73cb953bca30ed83e4505dd820edd9b5edffd28b`: full workspace `pnpm -r typecheck`, `pnpm build`, and `pnpm check:token-gates` pass locally. - All 422 focused connector/settings/shared-contract/migration tests and all 156 database-backed AI connection, hiring, reconnect, and durable-routing cases pass (69 hiring cases rerun after the final auth-precedence fix). The merged shared contract retains connection instructions and pool metadata. The pool migration is generated at sequence 0299 after the latest upstream migrations; this PR makes no lockfile changes. - All four full-app Playwright tests pass on the final head after a cold restart and migration, against the installed private plugin and isolated database, with no route or pool-API mocks. They cover hidden routing controls after manual opt-in, native pool creation, ordering, membership edits, rename, paused defaults, enabling/save/refresh persistence, stale edits, cancellation/removal, preserved underlying accounts, unavailable routers, and Used by avatars and profile links. Exact command: `PAPERCLIP_CONNECTION_POOL_E2E=1 AI_CONNECTIONS_TEST_COMPANY_ID=a37b9625-5ecf-4e29-8081-04df3d6e7d6f AI_CONNECTIONS_TEST_URL=http://127.0.0.1:3108 pnpm exec playwright test --config tests/ai-connections-app/playwright.config.ts connection-pools.spec.ts`. - [Native setup, ordering, and management screenshots](https://github.com/paperclipai/paperclip/pull/14967#issuecomment-6006976278) address the review follow-up. [Earlier selector, quota, and run-detail screenshots](https://github.com/paperclipai/paperclip/pull/14967#issuecomment-5971537316) show the runtime surfaces. Core previews: `pnpm --filter @paperclipai/ui storybook`, then **Connectors / Pool host** or **AI Connections / Connection pools**. Cloud owns its host-backed plugin stories; both repositories’ Operator Setup Required story assertions pass. - Live acceptance used OpenAI/Codex and Anthropic/Claude ACPX, resumed both exact sessions after restart, preserved pinned accounts through explicit reset and controlled quota deferral/recovery, and committed only two allocations across fourteen runs. A later UI-created task test again rotated OpenAI then Anthropic and resumed OpenAI through follow-up/restart/quota recovery. That later Anthropic execution was blocked by its saved OAuth token expiring (provider 401). No live usage probes ran. - The full local `pnpm test:run` was attempted earlier and did not complete because of macOS embedded PostgreSQL bootstrap/shared-memory failures and the 40,000-file Git fixture timeout. The focused database suites above now pass; full-suite verification is provided by the split CI lanes. The preceding CI run had one runtime readiness timeout; it passes locally both alone and inside the larger runtime suite. That larger local suite also encountered an embedded PostgreSQL setup failure and two macOS temporary-path alias assertions; those two assertions pass with canonical TMPDIR=/private/tmp. All final-head CI checks are terminal green, including full general/serialized server suites, Runner checks, browser E2E shards, canary verification, build, and typecheck. Greptile is 5/5 on that exact head with no unresolved threads. ## Risks - The migration adds routing tables and a credential epoch column. Install the private plugin only with the compatible Core contract. - Routing and each pool require opt-in. Production distribution and fleet defaults remain unchanged. - Unknown usage stays eligible. Known pinned exhaustion waits; revoked access requires operator repair. - Legacy adapters require compatible members. Runner model and effort overrides remain limited by qualified backend support. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository editing, code execution, and browser testing. The exact deployment model ID and context window 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 - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass (targeted suites; full-suite limitations are reported above) - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 20:44:37 -05:00
DottaandPaperclip	72ff3a9f27	Measure native tool context and expand bounded workflow evals (#15218) ... ## Thinking Path > - Paperclip manages persistent agents and their assigned work. > - Agents receive both fixed instructions and tool definitions. > - Moving a procedure into a tool description still adds model context. > - We need to measure the complete delivered catalog and test real outcomes. > - The tested reductions saved little space and introduced failing outcomes. > - This PR keeps measurement, bounded eval coverage and the original evidence. > - Production prompts, tools and runtime behavior stay unchanged. ## Linked Issues or Issue Description Refs: #15151, #14961, #14948, #14985. This adds the measurement and eval coverage needed to assess further native instruction changes. The attempted hiring and dependency reduction failed qualification and is excluded from the final diff. ## What Changed - Measure the actual standard-mode tool authority, including all 39 tools and their input schemas. Capture scripted native start, resume and continuation payloads and the OpenCode MCP declaration list. - Add OpenCode to the two explicit-only local hiring/reuse and delegation/feedback stories. Preserve the original task requests and independent oracles. - Apply one attempt per selected story and explicit company and lead-agent budget stops. - Select managed hiring credentials from the requested profile, including OpenRouter. - Run the existing Node test files under Node instead of collecting them as Vitest suites. - Retain sanitized comparison reports, original failed grades, source hashes and evidence gaps. ## Verification Final source: `2e7cef78eef7cdfe02265e0dcb03e855b8e50bd8`. Local verification passes: - Full repository `pnpm -r typecheck` and `pnpm build`. - Eval-support unit tests: 1,252 Vitest checks and 128 Node checks. - Eval TypeScript check and six final-source measurement tests. All 36 normalized components across nine scripted deliveries and the OpenCode MCP catalog match the baseline exactly; the complete standing projection is 49,200 bytes. Fresh review of this exact head is [5/5 with no remaining findings](https://github.com/paperclipai/paperclip/pull/15218#issuecomment-5996723170); all three review threads are resolved. [Final-head CI](https://github.com/paperclipai/paperclip/actions/runs/37354539126) passes all 47 jobs, including repository typecheck, build, tests, runner checks, browser shards and the canary check. One initial annotation-test timeout is retained in attempt 1; its seven-test suite passed locally, and the affected CI lane passed on one targeted retry. No source or paid eval rerun was needed. Both ready-transition security scans passed with zero annotations. The PR is out of draft and conflict-free. GitHub still requires code-owner approval for the `package.json` test-script change; its requested reviewers are already set. A byte-for-byte comparison against master context `a65ca0950834a85bb93bcc4b4042ecacdebfef53` confirms no production changes under `packages/` or production `server/` paths. The only server addition is a measurement test. No new paid rerun is needed to compare unchanged production bytes. This does not claim that existing product defects have been fixed. The rejected corrected experiment had baseline **5 PASS / 1 FAIL** and candidate **3 PASS / 3 FAIL**, including **two newly failing pairs**. The later readiness experiment had candidate **3 PASS / 3 FAIL** and baseline **3 PASS / 2 FAIL / one setup cell without a behavioral grade**. Its five comparable pairs had two new failures, two new passes and one unchanged pass. The missing baseline Codex cell never reached its provider step because Docker setup timed out. The observed missing behaviors include parent continuation, waiting for the latest child revision, revised ZIP delivery and OpenCode credential persistence. Those failures remain failures. Source review and passing CI do not regrade them. The original reduction saved 460 bytes; its first repair saved only 125 bytes, and the larger unqualified runtime repair increased the full projection. None of those production changes is shipped here. Read the [report](https://github.com/paperclipai/paperclip/blob/2e7cef78eef7cdfe02265e0dcb03e855b8e50bd8/doc/plans/2026-10-05-native-procedure-guidance.md) and its linked sanitized receipts for exact sources, original campaign links, pair-level results and evidence limitations. ## Risks - Full-catalog bytes are not model tokens, invoices, private vendor prompts, lazy-loading behavior or truncation proof. - The two stories are explicit-only and do not prove general coding quality or arbitrary resume behavior. Single trials do not establish causation or performance trends. - The retained OpenCode candidate credential guard failed. Cleanup removed the original provider database, so the precise persistence mechanism remains unknown. The guard is unchanged. - ACPX provider-execution IDs lack proven host-call mapping. No extra-work or feedback-consumption claim is inferred by matching names, order or counts. - PostgreSQL cannot start locally while the host's shared-memory slots are exhausted. Hosted CI must supply the full database checks; the full local database suite is not claimed green. ## Model Used OpenAI Codex, based on GPT-6, with repository tools and code execution. The exact deployment model ID and context-window limit 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>	2026-10-05 20:03:35 -05:00
Devin FoleyandPaperclip	efac8ff2f4	Add bounded Git integration restore diagnostics (#15291) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Sandbox runs must restore their workspace before finalization can succeed. > - Restore diagnostics identify the failed phase and operation. > - A Git integration exit code can still describe several different failures. > - This pull request adds fixed command labels and supported failure classes. > - Operators can distinguish these failures without collecting private Git output. ## Linked Issues or Issue Description Refs #15005. This branch includes merged #15268. It labels that change's locked ref transaction and nested branch probe without changing their behavior. **What happened?** A failed Git integration can report only `git_integration`, `unknown`, and an exit code. That evidence does not identify the failed command. Some Git versions also return exit 1 for both a merge conflict and an invalid object. **Expected behavior** Record a fixed command family and a supported failure class. Keep unknown cases as `unknown`. Exclude command arguments, process output, paths, repository URLs, filenames, and ref names. **Steps to reproduce** The tests create local repositories with conflicting commits, a missing object, and an expected-old ref mismatch. They call the real Git operations and inspect the resulting diagnostic. No hosted workspace or external provider is used. **Paperclip version or commit** Base: `e99854249c`. **Deployment mode** Built from source. The diagnostic applies to sandbox workspace restore. ## What Changed - Label Git integration calls with a closed command enum. Preserve arguments, options, errors, and retry behavior. - Recognize supported object and ref errors and OS permission codes. Require both exit 1 and completed tree output for a merge conflict. - Carry the closed fields through the existing restore receipt, saved adapter result, and Sentry projection. Revalidate saved metadata before projection. - Test nested wrappers, parallel failures, reused errors, handled probes, result settlement, and privacy with the real Sentry SDK. - Document the field contract and its limits. ## Verification - Eight focused suites pass: 295 tests. These cover Git sync, restore diagnostics, result settlement, teardown, sandbox runtime, failure projection, the real Sentry SDK, and native warm-workspace Git history. - `PAPERCLIP_REQUIRE_SENTRY_TEST_SDK=1 pnpm exec vitest run packages/adapter-utils/src/workspace-restore-diagnostics.test.ts packages/adapter-utils/src/workspace-restore-result.test.ts packages/adapter-utils/src/git-workspace-sync.test.ts packages/adapter-utils/src/workspace-restore-teardown.test.ts packages/adapter-utils/src/sandbox-managed-runtime.test.ts server/src/services/__tests__/run-failure-diagnostics.test.ts server/src/__tests__/run-failure-sentry-real-sdk.test.ts server/src/__tests__/native-workspace-sync-history.test.ts` - Local checks use Node 24.21.0 and the pinned pnpm 9.15.4. The optional Sentry SDK is pinned to the declared 10.71.0 and uses an in-memory transport. - `pnpm -r typecheck` and `pnpm build`: pass on the rebased head. - The full local `pnpm test:run` was stopped before source changes when #15268 merged and required a rebase. It reported three pre-existing company-skill cache test failures on macOS. The same three cases fail on clean bases `2c43b39167` and `e99854249c` and the earlier head with `EACCES` when publishing a read-only cache directory. The relevant test, service, and cache source blobs are identical. No cache changes are included. Later local test groups were not reached. - All 54 checks pass on exact head `8448896bc6`, including the post-ready security scan, with two intentional Storybook skips. Greptile scores this head 5/5 with no review threads. Full Linux CI covers the local groups that were not reached. - Independent review found no blocker. Its diagnostics, Git workspace, and native history suites pass 115/115 on this head. - A source comparison confirms that removing only diagnostic wrappers yields the merged upstream Git integration code exactly, including ref locks, transaction protocol, arguments, options, and retry decisions. - The clean base has stale dependency overrides in its lockfile. Local installation resolved them as the existing PR CI fallback does. No manifest or lockfile change is included. - `git diff --check` and a local secrets/PII review pass. ## Risks A Git version or localized message may not match a known failure form. Such cases remain `unknown`. Up to 16 KiB of stderr and the bounded tree-ID prefix of stdout are inspected only in memory. The saved fields contain enum values only. These diagnostics do not establish workspace recovery or authorize retries. Restore decisions and Git mutations are unchanged. No schema migration is required. ## Model Used OpenAI Codex, GPT-6, with reasoning, tool use, and code execution. The service does not expose the exact model deployment 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 #` / `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>	2026-10-05 17:10:05 -07:00
Devin FoleyandPaperclip	e99854249c	fix(workspaces): restore rebased sandbox history against its starting snapshot (#15268) ... ## Thinking Path > - Paperclip manages agents and preserves their work across runs. > - Sandbox execution restores Git history and files to the host workspace. > - An agent can rebase or amend its branch before it finishes. > - Restore currently treats the original and rewritten tips as concurrent work. > - That merge can conflict even when the host has not changed. > - This change uses the starting Git snapshot to accept rewritten history safely. ## Linked Issues or Issue Description **What happened?** A successful sandbox turn can end with `workspace_restore_failed` after the agent rebases and pushes its branch. The restore step merges the original host tip with the rewritten sandbox tip. This can recreate conflicts that the agent already resolved. **Expected behavior** Accept the rewritten history when the host still has the recorded starting branch and commit. Preserve concurrent host work through the existing merge and recovery paths. **Steps to reproduce** 1. Start a sandbox from a feature branch. 2. Rebase that branch onto an upstream commit that changes the same file. Resolve the conflict in the sandbox. 3. Restore the sandbox while the host remains on the original commit. 4. The old implementation attempts a conflicting Git merge and fails the run after the agent finishes. **Paperclip version or commit** Reproduced on `cab4263dc9` with a real Git rebase fixture. **Deployment mode** Self-hosted server with sandbox execution. Related work: #15005 records restore failure stages. #11638 preserves unrelated imported history with a graft. #10601 handles bundle prerequisites. Those changes do not distinguish a rebase from a concurrent host edit. ## What Changed - Pass the run's starting Git branch and commit into sandbox history integration. - Adopt a related rewritten tip when the host still matches that snapshot. Keep the expected-old-value ref update and bounded retry. - Verify the branch attachment inside a prepared Git transaction while Git holds its ref locks. Abort if a checkout changed the branch. - Check host and sandbox Git identity before warm reuse, including nested repositories. Restage when their tips or branches differ. - Reject host branch changes and unrelated imports after a concurrent host commit. - Retain the existing unrelated-history graft for an unchanged host and the conservative behavior for callers without a snapshot. - Export a full bundle for an intentional reset to an ancestor so restore receives the actual sandbox tip. - Add real Git and sandbox restore regressions. Document the restore contract. ## Verification - Eight Git sync, sandbox restore, and native workspace suites pass: 255 tests. - The new checkout-race and warm-reuse regressions failed before the fixes. Real Git hooks verify that prepared transactions prevent a concurrent HEAD change. - `pnpm -r typecheck` passed after rebasing onto `984f092ddf` and applying the Apex findings. - `pnpm build` passed on `f5132603d6`. - The earlier `pnpm test:run` attempt reported three `company-skills-service.test.ts` failures on macOS (`EACCES` renaming a read-only staging directory). The same failures reproduced on unchanged master. The broader run was stopped after confirming that baseline failure; it was not a full-suite pass. Those source and test files are unchanged in the current base. - [Apex review](https://github.com/paperclipai/paperclip/pull/15268#issuecomment-6002549219): 5/5 on `f5132603d6`, requested with `@greptileai apex review`. All three historical findings are addressed and all review threads are resolved. - All CI gates passed on `f5132603d6` (53 successful checks, two skipped). The signoff-policy browser fixture initially timed out waiting for a local process-agent run; its one retry passed without code changes. [CI run](https://github.com/paperclipai/paperclip/actions/runs/37387511152) ## Risks - A recorded starting snapshot now authorizes replacement of related rewritten history. A stale or changed host tip retains the concurrent-history path. Ref writes still compare the expected old commit. - Branch changes and unrelated rewrites after host advancement require recovery instead of replacing host work. - An intentional reset to an ancestor uses a full bundle. Large histories can increase transfer time in that case. - The existing directory merge rules remain in force. The fix does not resolve an earlier failed restore or replay its external actions. ## Model Used OpenAI GPT-6 via Codex, with reasoning, repository analysis, code editing, and local test execution. The exact deployment model ID and context window 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 — 255 affected tests; the broader-suite baseline failure is documented above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 16:43:15 -07:00
DottaandPaperclip	4857799a88	feat(connections): deliver saved instructions to authorized agent turns (#15216) ... Persist optional connection instructions and deliver authorized snapshots to agent execution prompts. Keep provider templates with each app definition, preserve edits and opt-outs, and replace sessions when guidance or access changes. Use shared production settings across setup and Permissions, with source visibility in agent Instructions. Add the initial memory-provider defaults and managed Honcho workspace configuration. Include migration 0298 and regression coverage for generic providers, runtime delivery, authorization, and catalog regeneration. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-10-05 18:16:00 -05:00
DottaandPaperclip	2c43b39167	refactor(ui): share task composer and project/worktree controls (#15091) ... Use the shared composer for new tasks, including project and worktree selection, rich mentions and slash commands, and remembered task settings. Save project preferences after successful task updates. Show known agent models by name and use Default for unknown runtime defaults. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-10-05 18:06:40 -05:00
Nicky LeachandPaperclip	984f092ddf	Add project and agent resource lifecycle hooks (#15280) ... ## Thinking Path > - Paperclip manages agents and projects for work. > - Plugins need reliable lifecycle hooks for these resources. > - In-process notifications can disappear during a restart. > - A lifecycle record must commit with the resource change. > - This pull request adds a generic lifecycle journal without provider calls. > - A later plugin delivery layer can use these records without losing lifecycle changes. ## Linked Issues or Issue Description **Problem or motivation** Resource provisioning needs durable hooks for agent hiring, pause, resume, termination, and project creation. Hooks must cover shared service callers, budget actions, and approval paths. Capture should work for all plugins and deployment modes. **Proposed solution** Record content-free events in the resource transaction. Deduplicate creation and unchanged status. Preserve each pause/resume cycle. Commit approval, activation, and creation together. Commit termination and API-key revocation together. **Alternatives considered** Event subscriptions alone cannot survive process failure. Cloud-only capture would exclude other plugins. Provider calls inside tenant transactions would couple resource creation to external services. **Roadmap alignment** This is lifecycle infrastructure for the existing plugin system. It adds no provider, adapter, UI, or plugin read API. Repository mutations and backfill remain separate work. Searches found no duplicate lifecycle-journal PR. ## What Changed - Add a company-scoped lifecycle journal and an additive migration. - Record hired-agent and project creation from shared services. - Record agent pause, resume, and termination, including generic updates and budget actions. - Lock agent state changes to suppress concurrent duplicate hooks. - Make hire approval, rejection, and termination transactions atomic with their lifecycle records. - Document capture, ordering, future per-plugin acknowledgments, and migration scope. ## Verification - Full workspace typecheck: `pnpm -r typecheck` passed. - Production build: `pnpm build` passed. - Focused agent, project, approval, budget, and built-in regressions: 102 tests passed across 9 suites. - Final lifecycle journal check: 14 tests passed. It covers repeated/concurrent transitions, rollback, key revocation, self-hosted capture, and company boundaries. - Review regression: 32 lifecycle and approval tests passed, including rejected-hire rollback and retry; server typecheck passed after the fix. - The initial local `pnpm test:run` overlapped the rejection fix and reported the new rollback regression against the earlier service code. A fresh lifecycle run passed all 14 tests. Fresh full local shards were stopped once the complete CI test matrix passed on the final commit. - `git diff --check` and a local secret/PII scan passed. - Greptile: 5/5 on `3ee3903f8e1a171ea3ba2bea9caa2b5766a3f807`, with the review thread resolved. - [Complete CI passed](https://github.com/paperclipai/paperclip/actions/runs/37374227222) on `3ee3903f8e1a171ea3ba2bea9caa2b5766a3f807`: general server/chat/workspace tests, serialized server suites, runner checks, build, typecheck, canary, and all end-to-end shards. The requester waived CI during the Actions outage, but the workflow subsequently completed successfully. ## Risks - The migration creates an empty table. It does not scan or backfill existing resources. - Apply the normal database migration before running this server version. Event-write failure intentionally rolls back the resource change. - Pending hires cannot bypass approval through pause or resume. - Capture works on all deployments. Plugin delivery, retention, retries, and provider actions remain separate work. No plugin can read this journal through a new API in this PR. - Future delivery must enforce company scope, track acknowledgments per plugin, and preserve resource order. A global sequence cursor can skip uncommitted transactions. - A termination hook does not authorize deleting persistent volumes. ## Model Used OpenAI Codex, based on GPT-6, with reasoning, repository inspection, code execution, and tool use. The exact deployment model ID and context window 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>	2026-10-05 15:58:38 -07:00
4 people	6c36c07a4f	feat(adapters): add GPT-6.1 Sol and refresh shared coding harness pins (#14942) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents run through coding-agent adapters and the native runner. Both use the same installed provider CLIs, model catalogs, and reasoning controls. > - OpenAI released GPT-6.1 Sol (`gpt-6.1-sol`) in Codex. Anthropic released Claude Sonnet 5.5. The static Codex, Bedrock, and OpenCode catalogs do not list these IDs. > - The shared provider pack pins Codex 0.156.0 and OpenCode 1.18.32. The evaluation image pins older Grok, Gemini, Kimi, Cursor, and GitHub CLI releases. Codex 0.156.0 has no bundled metadata for GPT-6.1 Sol. > - A model entry without a current harness, or a harness pin without its runner integrity checks, fails at run time. > - This pull request adds the verified model IDs and moves the harness pins, executable digests, controller checks, and image pins together. > - The benefit is that operators can select the current models, and the native and local adapters share one current CLI installation. ## Linked Issues or Issue Description Refs #13829 and #13838 (the September 22, 2026 model and harness refresh). Related pull requests: #14993 (merged October 5, 2026, superseding #14816) added the direct Claude Sonnet 5.5 entry and refreshed the Claude runtime to Agent SDK 0.3.286 / Claude Code 2.1.286. This pull request does not change the Claude runtime or the direct Claude model list; it keeps the #14993 pins and adds only the Bedrock Sonnet 5.5 ID. After #14993 merged, this branch was rebased onto `master` (October 5, 2026). The six overlapping pin regions (`docker/daytona-runner/Dockerfile`, `docker/daytona-runner/README.md`, `package.json`, `pnpm-workspace.yaml`, `packages/adapters/claude-local/src/index.test.ts`, `packages/paperclip-runner/src/backends/native-backend-factory.test.ts`) were resolved by keeping this pull request's Codex 0.160.0 and OpenCode 1.18.34 pins next to #14993's Claude 0.3.286 / 2.1.286 pins, taking the union of the Sonnet 5.5 model IDs in the Claude test, and merging both README paragraphs. The Sonnet 5.5 effort and CLI-gate lines in the Claude adapter were identical in both pull requests and merged without a diff. #14917 and #14918 reordered the Claude and Codex model lists earlier; the new entries sit where those ordering rules put them. Sources checked on 2026-10-02: - [OpenAI Codex models](https://learn.chatgpt.com/docs/models): GPT-6.1 Sol uses `gpt-6.1-sol`, supports reasoning efforts from Light to Ultra, and has Standard and Fast modes at launch. The page also records that `gpt-5.4` and `gpt-5.4-mini` retired from Codex with ChatGPT sign-in on August 31, 2026, and that `gpt-5.5` retires on October 14, 2026. Neither retirement applies to the OpenAI API. - [Codex CLI releases](https://github.com/openai/codex/releases) 0.157.0 through 0.160.0. The bundled model metadata in the 0.160.0 Linux binary contains `gpt-6.1-sol`. - [Claude Sonnet 5.5](https://platform.claude.com/docs/en/models/sonnet-5-5/overview): Bedrock ID `anthropic.claude-sonnet-5-5`, released September 28, 2026. - [OpenCode releases](https://github.com/anomalyco/opencode/releases) 1.18.33 and 1.18.34 (fixes only). The OpenCode model registry lists both added provider-qualified IDs. - npm `latest` tags for `@xai-official/grok` 1.0.46, `@google/gemini-cli` 0.62.0, and `@moonshot-ai/kimi-code` 2.1.1. [xAI](https://docs.x.ai/docs/models), [Google](https://ai.google.dev/gemini-api/docs/models), and [Kimi](https://www.kimi.com/code/docs/en/kimi-code/models.html) list no newer coding models. - Cursor CLI 2026.10.01-e373342 is the version the official installer resolves. The pinned digest is the SHA-256 of the versioned Linux x64 archive. - [GitHub CLI 2.102.0](https://github.com/cli/cli/releases/tag/v2.102.0) (security fixes). The pinned digest matches the release `checksums.txt`. ## What Changed - Codex adapter: add `gpt-6.1-sol` to the model list, the Fast mode list, and the Ultra effort set. It is the first entry: #14918 orders the list newest version first, and its description notes the ChatGPT app lists GPT-6.1 Sol first. Update the adapter documentation text. - Claude adapter: add `us.anthropic.claude-sonnet-5-5` (Bedrock Sonnet 5.5) to the Bedrock catalog in the newest-Sonnet slot after Opus 5.5; `us.anthropic.claude-sonnet-5` moves into the older-Sonnet group, matching what `sortClaudeModels` from #14917 produces at runtime. Any Sonnet 5.5 ID (direct or Bedrock-qualified) now gets the documented `xhigh` and `max` efforts and requires Claude Code 2.1.284 or later on the CLI lane (the Claude Code changelog entry for 2.1.284 adds `claude-sonnet-5-5`). These two lines are identical to the ones #14993 merged, so the branch carries no diff for them. - OpenCode adapter: add `openai/gpt-6.1-sol` and `anthropic/claude-sonnet-5-5` to the static fallback catalog. - Codex runtime pin 0.156.0 → 0.160.0 in the root and workspace overrides, the runner package, the Codex ACP package patch, the qualified ACPX profiles, the Linux x64 executable digest, the Rust provider backend and its tests, the provider-pack manifest pins, the remote controller pins, the sandbox npm install spec, and the opt-in qualification scripts. - Remote Codex compatibility window: upper bound 0.157.0 → 0.161.0. The minimum stays at 0.149.0. - OpenCode runtime pin 1.18.32 → 1.18.34 in the runner package, the materialization script, the server and Rust qualified versions, the eval and live-session labels, fixtures, and the configuration label. - Evaluation image (`docker/daytona-runner/Dockerfile`): Grok CLI 1.0.46, Gemini CLI 0.62.0, Kimi Code 2.1.1, Cursor CLI 2026.10.01-e373342 with its digest, GitHub CLI 2.102.0 with its digest, Codex and OpenCode version probes, and the refreshed lockfile digest. The Claude Code 2.1.286 probe comes from #14993 and is unchanged here. - `pnpm-lock.yaml` is not part of this pull request. The repository's pull request gate rejects lockfile edits, and the refresh bot regenerates the lockfile on master (the same flow #13838 used). The Dockerfile `PAPERCLIP_RUNNER_LOCK_SHA256` default is the digest of the lockfile that `pnpm install --resolution-only --ignore-scripts --no-frozen-lockfile` (the refresh workflow's command) produces for the combined pins on the rebased branch (`e1856797…`); that lockfile differs from master only in the `@openai/codex` 0.160.0 platform packages, the `@anthropic-ai/claude-agent-sdk` 0.3.286 override that #14993 introduced (the open refresh-bot pull request #14872 carries that part), `opencode-ai` 1.18.34 with its Linux x64 baseline, and the `codex-acp` patch hash. - Documentation: runner README, runner compatibility doc, environment variable example, and a new `doc/adapter-model-audit-2026-10-02.md` with sources and deferred items. - Tests: Codex adapter catalog, server adapter models, Codex compatibility window, native session executor pins, runner package contract, OpenCode materialization, and UI effort options. Unchanged on purpose: Claude Agent SDK 0.3.286 / Claude Code 2.1.286 (already on `master` from #14993), ACP bridges (`acpx` 0.13.1, `claude-agent-acp` 0.73.0, `codex-acp` 1.6.2; newer upstream releases need a separate qualification), the native Grok runtime 1.0.13, Pi 0.84.2 / 0.87.1 (the Pi 1.0 runner stack covers it), and Hermes 0.19.0 (current). `gpt-5.4` and `gpt-5.4-mini` stay in the picker because the OpenAI API still serves them. ## Verification Run on Linux x64 with Node 25.9.0 and pnpm 9.15.4 after `pnpm install --no-frozen-lockfile` (the refreshed lockfile stays local; see above). The results below were re-run on the rebased head (October 5, 2026) for the suites the conflict resolution touches; the other rows are from the original run and are covered by CI on every push: - Rebased head: `packages/adapters/codex-local` 482 passed; `packages/adapters/claude-local` 340 passed, 4 failed (`execute.remote`, `test.probe`, `execute.acp-fallback`, `acp` spawn/env-hardening cases that fail identically on unchanged `master` in this host environment); `server` adapter-models + codex-runtime-compatibility + native-session-executor + adapter-registry 607 passed, 1 failed (the same adapter-registry override-pause case as before, also failing on `master` here); `packages/paperclip-runner` native-backend-factory + qualified-profiles 36 passed; `ui` codex-reasoning-effort + config-fields + model-utils 19 passed. Rust, full typecheck, build, and the Docker image are left to CI as before. - `vitest run` in `packages/adapters/codex-local`: 13 passed. `vitest run` in `packages/adapters/claude-local` (whole package, including the new Sonnet 5.5 gate and effort tests): see the latest CI run and the comment below. `vitest run` in `packages/adapters/opencode-local`: 48 passed, 1 failed (`runtime-config.test.ts` reads the host `PAPERCLIP_OPENCODE_PROVIDERS` variable; it fails the same way on the unchanged base). - `vitest run src/__tests__/adapter-models.test.ts src/services/native-runtime/codex-runtime-compatibility.test.ts src/__tests__/adapter-registry.test.ts` in `server`: 84 passed, 1 failed (`adapter-registry.test.ts` override pause test; it fails the same way on the unchanged base). - `vitest run` in `ui` for `codex-reasoning-effort`, `agent-setup-fields`, `config-fields`, and `ComposerRunSettingsPicker`: 25 passed. - `node --test test/acpx-codex-package-contract.test.mjs scripts/materialize-opencode-binary.test.mjs scripts/runner-protocol-eval-campaign.test.mjs` in `packages/paperclip-runner`: 23 passed. The package contract test verifies the installed Codex ACP executable digest and the 0.160.0 patch pin. - `vitest run src/drivers/acpx src/backends src/drivers/opencode src/live/live-session.test.ts` in `packages/paperclip-runner`: 626 passed, 5 failed, 1 skipped. The 5 failures (`installation-integrity.test.ts` `/proc/self/fd` module loading and one OpenCode answer-selection test) also fail on the unchanged base under Node 25; Linux CI runs Node 24. - `pnpm run test:opencode:qualification` in `packages/paperclip-runner` against the installed OpenCode 1.18.34 executable: passed. - `codex --version` from the installed pack prints `codex-cli 0.160.0`. The Linux x64 executable digest `12eb3e81…652aad` was computed from the `@openai/codex@0.160.0-linux-x64` archive after checking its registry `dist.integrity`. - `pnpm check:token-gates`: all gates clean. - `pnpm run typecheck:typescript` in `packages/paperclip-runner`: passed. Package typechecks ran one at a time; see the comment below for the server and UI results. Not run here, and needed from CI: - Rust tests and `pnpm -r typecheck` / `pnpm build` for the server (no `cargo` in this environment; the server typecheck prepares the runner vendor build). - The Docker evaluation image build and the real-binary Codex startup and session-resume probes (no Docker; the probes need the compiled `paperclip-runnerd`). The trusted CI runner workflow covers them. - Authenticated inference with any new model. This change is metadata and startup validation only. ## Risks - Codex 0.160.0 changes the bundled model catalog and app-server behaviour (authoritative provider catalogs, incremental running-turn tracking). The patched `codex-acp` 1.6.2 bridge is unchanged and declares `^0.148.0`; it worked with 0.156.0 under the same override. If CI probes show a protocol change, the pin can return to 0.156.0 by reverting this pull request. - The compatibility window upper bound moves to `<0.161.0`. Remote images with Codex 0.157 to 0.160 become accepted. Older images stay accepted down to 0.149.0. - Until the refresh bot lands the regenerated lockfile on master, the Dockerfile lockfile digest default does not match the committed lockfile. The trusted CI workflow computes the digest from its own resolution at build time, so this affects only a local build that passes no digest. - Existing saved model selections and effort settings are not changed. Agents on `gpt-5.4` or `gpt-5.5` with ChatGPT sign-in need a model change before the OpenAI retirement dates; that is documented, not enforced. - Rollout order: deploy the controller and runner from this change before promoting a sandbox image that carries these pins. Older controllers reject the new provider-pack pins. ## Model Used - Claude Fable 5.1 (Anthropic, model ID `claude-fable-5-1`), 1M context window, adaptive thinking, tool use. The model ran as a Paperclip agent through the Claude Code harness, performed the web research, edited the code, and ran the tests listed above. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: Bender (Fable) <noreply@paperclip.ing> Co-authored-by: devinfoley <139239+devinfoley@users.noreply.github.com> Co-authored-by: Claude Fable 5.1 <noreply@anthropic.com>	2026-10-05 13:59:08 -07:00
DottaandPaperclip	b43073d11f	feat(connections): sync and group accounts managed by aggregators (#15254) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Connections give agents governed access to external tools. > - Aggregator gateways can expose accounts that users already connected upstream. > - The Apps catalog did not show those accounts or their current provider status. > - Separate cards and setup tasks also made account ownership unclear. > - This pull request discovers upstream accounts and groups them under one app card. > - Users can find connected apps while each provider keeps control of its accounts. ## Linked Issues or Issue Description **Subsystem affected** Connections across the database, shared contracts, server, and board UI. **Problem or motivation** Users cannot see which apps are connected through a saved aggregator gateway. Native and upstream accounts need one app card. Discovery must preserve company, user, gateway, and credential boundaries. **Proposed solution** Sync account metadata from Composio, Arcade, and supported Executor gateways. Keep upstream account management in each provider. Use source chips and search to browse the catalog. Preserve native setup and the gateway's existing access policy. **Alternatives considered** Creating a local executable connection for each upstream account would duplicate authorization state. Using an agent task for routine Composio setup would add an unnecessary step. The board now calls the saved gateway directly for that setup. **Roadmap alignment** This extends the shipped Connected Apps and MCP Tool Gateway features in ROADMAP.md. The duplicate search found no open PR for managed account discovery. Related work: Refs #13755, Refs #13941, Refs #14725, Refs #13855. Open PR #12906 covers adjacent toolkit routing work. ## What Changed - Add provider-neutral discovery, sync, and refresh APIs. Preserve the Composio API paths. - Cache observations by company, saved gateway, viewing user, and credential version. Retain stale observations after failed or incomplete scans. - Add optional Arcade account sync credentials in the vault. Discover Executor accounts through its supported inventory interface. - Group native and upstream accounts in one app card. Imported account menus open their provider. Gateway menus own refresh and sync setup. - Add Paperclip, Composio, Arcade, Installed, and All chips. Show 50 catalog entries per page. Keep connected accounts above discovery. Keep explicit provider searches scoped. - Simplify Composio app setup and refresh its connected app list on the gateway Permissions page. - Add a compact agent access card and task creation defaults for connection setup. Preserve explicit blocks and approval policies. - Add two replay-safe migrations, service and UI tests, Storybook journeys, and acceptance stories. ## Verification - Passed the repository typecheck, full build, token gates, and migration ordering check. - Passed the focused provider adapter, connection interaction, and catalog tests after rebasing onto master. - Passed all nine database sync and migration replay tests using a disposable database on the test-drive PostgreSQL cluster. Removed that database after the run. - Verified Arcade cursor pagination against its official Go SDK and passed all eight adapter tests, including short and incomplete pages. - Passed all 45 interaction tests after making the exact requested tools and their Allowed/Ask first permissions visible before granting access. Verified the compact card in Storybook. - Passed the complete UI suite on the final code: 683 files and 7,432 tests, including the corrected Composio destination assertions. Passed 130 focused tests for the UUID, management-link, and health-status corrections. - Passed 22 Composio setup/sync tests, 23 connection-intent service tests, and the connection migration test in separate disposable databases. Database startup alone was substituted; the suites exercised their real SQL and services. - Passed all 10 OpenAPI route checks and the full-stack connection-intent browser test, including scoped consent, agent continuation, and task completion. - The local full runner encountered embedded PostgreSQL startup failures on this loaded macOS host. The earlier in-flight run also held the pre-fix Arcade transform; a fresh run of the final provider suite passes. The final-head CI is queued during GitHub’s active Actions incident: https://www.githubstatus.com/. The previous run also lost several runners simultaneously; its real catalog assertion failures are fixed and the fresh complete UI suite passes. - Tested the real test-drive server in the embedded browser with a live Composio gateway. Detected Airtable and Circleback. Verified refresh progress, account rows, source chips, search scope, and 50-entry pagination. - Arcade and Executor coverage uses provider fixtures. Live credentials were unavailable. - Storybook builds successfully and includes grouped native/provider accounts, stale and unavailable discovery, optional Arcade setup, and mobile states. The acceptance document records the simulated and live coverage separately. - Greptile reviewed final commit `217b024c27b5933e773ce9419c4e92b1032042c6` at 5/5. All six review threads are resolved, security scans pass, and the PR has no merge conflicts. The outstanding remote checks are `ci / Select trusted runner` and `review`, queued by GitHub. They need to complete before merge. ## Risks - Provider response changes can break inventory discovery. Failed scans retain observations and show stale status. - Composio scans only the supported catalog and can take time. Large inventories run in the background with progress and a bounded lease. - Arcade requires a project API key and user ID when the gateway cannot supply them. This key is used only for discovery. - Executor discovery depends on the server's exposed inventory tools. Unsupported servers report unavailable discovery. - Cached account rows do not grant access or create executable connections. Gateway policies still govern tool use. Account deletion and per-app authorization remain upstream. - The migrations add tables and one nullable column. Replay preserves existing rows and company-scoped foreign keys. ## Model Used OpenAI Codex, based on GPT-6. The session does not expose a more specific serving model ID or context limit. Used reasoning, repository tools, code execution, and browser verification. ## 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 - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 14:57:04 -05:00
Devin FoleyandPaperclip	17fa718675	Retry transient disconnects in scoped read queries (#15248) ... Apply existing bounded transient-disconnect retries to audited dashboard, heartbeat-list, and base issue reads. Rebuild each query per attempt and preserve authorization, company scope, enrichment, and final error propagation. Validation: exact-head Greptile 5/5, passing CI, no unresolved review threads, and a clean merge. Detailed verification and limitations are recorded in the pull request. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-10-05 10:58:56 -07:00
Devin FoleyandPaperclip	3b47a6befd	Record bounded workspace restore failure stages (#15005) ... Capture allowlisted restore substep and failure metadata for future diagnosis while preserving workspace recovery behavior, error identity, cleanup ordering, and privacy. Validation: exact-head Greptile 5/5, passing CI, no unresolved review threads, and a clean merge. Detailed verification and limitations are recorded in the pull request. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-10-05 10:58:32 -07:00
Devin FoleyandPaperclip	2e67ea8dc7	fix: renew explicit continuation authorization for bounded retries (#14987) ... Preserve exact user-authorized continuation receipts across bounded retries and retain the task claim through owned retry admission. Stop, reassignment, superseding input, and active cleanup continue to block unsafe continuation. Validation: exact-head Greptile 5/5, passing CI, no unresolved review threads, and a clean merge. Detailed verification and limitations are recorded in the pull request. Co-Authored-By: Paperclip <noreply@paperclip.ing>	2026-10-05 10:57:57 -07:00
DottaandPaperclip	59015846ae	fix(chat): keep dismissed task questions in the feed (#15229) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents ask users questions in task chats and Agent Chat. > - Agent Chat keeps unanswered questions as compact entries in the feed. > - Regular task chats still show a pending composer badge after dismissal. > - A page reload can also open a dismissed question form again. > - This pull request applies the same feed behavior to both chat types and saves dismissal per person and task. > - Users can continue the chat and return to the original question later. ## Linked Issues or Issue Description **What happened?** Dismissing a question in a regular task chat leaves a pending composer badge. The question can return after a page reload. The earlier feed behavior only applied to Agent Chat. **Expected behavior** A dismissed question stays in the feed. Its form and pending badge leave the composer. Reload preserves dismissal. Opening the feed entry restores the original question and draft answer. **Steps to reproduce** 1. Open a regular task chat with a pending question. 2. Select an option, then dismiss the form. 3. Reload the page. Check that the composer stays clear. 4. Open the question in the feed. Check that the draft is restored. 5. Submit the answer. Check that the answered receipt appears. **Paperclip version or commit** Reproduced on master at `a386a599983519eb1d399f8b770bfccdb2a74762`. **Deployment mode** Browser UI in local and authenticated instances. This change does not depend on the agent adapter. Related work: Refs #14613, which added the Agent Chat feed behavior. Refs #9141, which validates real answers on the server. Refs #11434, which tracks comment-driven changes to interaction state. This PR changes question presentation only. ## What Changed - Show compact unanswered question entries in regular task chats. - Exclude durable questions from composer pending counts and navigation. - Save dismissal in local storage per person and task. Merge the latest saved IDs so dismissals from another tab survive reload. A new question can still open its form. - Keep the original question pending and answerable. Keep approval and permission controls. - Run the question-history regressions in both chat modes. Add reload, new-question, user/task scope, and stale-tab coverage. - Share the real-component Storybook fixture. Add a regular task test drive and an interactive dismissal/answer scenario. - Update the planning-mode browser test to check a dismissed question in the feed after reload and on mobile. - Update the design rules, implementation spec, and preview instructions. ## Verification - 329 focused thread, composer, and interaction-card tests pass. - `pnpm -r typecheck` passes. The UI typecheck also passes after the review fix. - `pnpm build` passes for the full repository. - UI build, Storybook build, and token gates pass. The UI build and token gates were rerun after the review fix. - Greptile gives final commit `98f71f221` a 5/5 score. The current-head check passes, and there are no unresolved review threads. - All 56 current-head checks are terminal: 54 pass and two conditional Storybook jobs skip. The CI run includes the full test shards, build, typecheck, browser tests, aggregate verification gate, and package canary. - The stale-tab regression fails in both chat modes before the review fix and passes after it. - `pnpm exec playwright test --config tests/e2e/playwright.config.ts tests/e2e/planning-mode-visual-verification.spec.ts` passes against a throwaway local server. It checks dismissal, reload, task navigation, and desktop/mobile planning controls. - The duplicate local `pnpm test:run` attempt was stopped after the full CI test suites passed. It did not complete locally. - Manual browser test: select Green, dismiss, reload, reopen, submit the saved answer, and inspect the answered receipt. Also send a new message while the unanswered question remains in the feed. The test drive uses real UI components with fixture response callbacks. - To repeat the browser test, run `pnpm storybook`. Open **Chat & Comments → Task Chat Unanswered Questions → Test Drive**. ## Risks - Dismissal is a browser-local preference. It does not sync to another browser or device. Clearing local storage removes it. - When local storage is unavailable, dismissal lasts for the mounted thread only. - Unanswered questions can accumulate in the feed. They stay pending until answered or resolved through the existing API. - No database migration, API change, or change to approval permissions. ## Model Used OpenAI Codex, `gpt-6.1-sol`, with xhigh reasoning, repository editing, code execution, and browser control. 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 `#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>	2026-10-05 10:55:38 -05:00
DottaandPaperclip	a65ca09508	fix(runner): settle accepted results after shutdown failures (#15217) ... ## Thinking Path > - Paperclip manages AI agents and the tasks they perform. > - The native runner saves tool results and completion reports before it releases a session. > - Large project discovery responses can exceed the durable command limit. > - A shutdown failure can leave a saved answer waiting for workspace repair. > - Recovery reused the old assessment for a different status decision, which violated a database constraint. > - This pull request bounds discovery responses and lets recovery commit the saved result after workspace repair. > - The benefit is a task that reaches its correct final status without another provider turn. ## Linked Issues or Issue Description **What happened?** A native run can save its final answer, fail during shutdown, and leave the task In Progress after workspace repair succeeds. Reconciliation tries to reuse the failed-workspace assessment for a new decision. The one-decision-per-assessment constraint rejects the write. Replaying the old decision can also retain a fresh coordinator lease. Separately, retained session cleanup only recognizes the old `adapter_failed` label. The project-list tool returns full project records, including large descriptions and workspace configuration. A large response exceeds the runner's durable command limit. The settlement diagnostic previously recorded only a failure flag. **Expected behavior** Project discovery stays within the command limit. Recovery finishes the saved result after workspace repair, releases its lease, and preserves the original error for inspection. It does not repeat provider work or relax session ownership checks. **Steps to reproduce** 1. Return large project records from `list_projects` and observe an oversized semantic result. 2. Persist an accepted native completion result, then record a shutdown failure. 3. Finalize with a failed workspace, repeat that attempt, then record successful workspace repair. 4. Reconcile the run. Before this fix, the issue stays In Progress. **Paperclip version or commit** Reproduced on `a386a599983519eb1d399f8b770bfccdb2a74762`. **Deployment mode** Self-hosted server with Paperclip Runner. Related transport work: #12208 drains queued events; #12241 resumes interrupted semantic calls. This change addresses bounded project discovery and accepted-result finalization. ## What Changed - Read bounded project summary projections from the database and return at most 50 authorized summaries with a continuation cursor and explicit description truncation. Agent and run trust boundaries narrow the database candidates; project-specific policies still receive full authorization. Only visible projects determine continuations. The default project-list API remains unchanged. - Record bounded, content-free settlement failure causes for command limits, storage errors, and rejected dispatch. - Include workspace state in assessment identity. Preserve the initial assessment for interrupted finalization, and commit replacement assessment and decision references together. - Release the coordinator lease when an existing decision is replayed, without repeating its effects. - Clear stale errors when recovery succeeds and retain them in `recoveredExecutionFailure`. - Accept both current and legacy close-failure labels in the existing exact-state cleanup path. - Add regression tests and update the tool contracts and recovery documentation. ## Verification - Red: the new project paging, settlement diagnostic, current cleanup label, and repaired-workspace regressions failed on the original implementation. - Green: protocol/catalog/tool checks (117 tests), the full project-tool and finalizer suites (47 tests), cleanup ownership cases (70 tests), and cleanup sweep cases (4 tests) pass. Database-backed pagination covers large descriptions/configuration, complete enumeration, uppercase cursors, agent/run restrictions, project-policy scope contributions, and identical results/cursors when hidden projects are added. - The initial CI failures in interrupted Board waits, contended endpoint proof, and semantic schema validation were reproduced and fixed; all affected cases pass locally. - `pnpm -r typecheck` — passed. - `pnpm build` — passed. - `pnpm test:run` — started before the review corrections; it spanned several source revisions and was stopped after reporting old-behavior and timing failures. It is not claimed green. Fresh project and recovery suites pass; the recovery suite also passes all 25 cases with the broad runner’s isolated home/config. The Slack timing case passed in isolation. Latest-head CI is the authoritative complete test matrix. - Existing authorization suite — 66 tests passed. - Latest-head CI on `8e5763d915aa6f375bdab6601996899ea01496fc` — 55 checks passed, 4 intentionally skipped, no pending or failing checks. - Greptile — 5/5, zero unresolved threads on the same commit. - `git diff --check` — passed. ## Risks - `list_projects` now returns summaries. Callers must follow `nextCursor` and use the authorized project API for full records. Candidate narrowing is only an optimization: project policy and responsible-user authorization remain authoritative. - Assessment identity changes for workspace finalization. Existing evidence remains intact; no migration is needed. - Cleanup still requires matching identities, settled tool evidence, and verified process ownership. Unknown tool outcomes remain blocked from session reuse. ## Model Used OpenAI Codex, based on GPT-6, with reasoning, code execution, and GitHub tools. The exact serving model ID and context window 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>	2026-10-05 10:31:11 -05:00
DottaandPaperclip	a386a59998	Reduce repeated native completion guidance and preserve final replies (#15151) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Native agents receive task constraints and completion tools from Paperclip. > - Completion tools already define the procedure for reporting a result. > - Repeated procedure text adds instructions to each full task turn. > - The final reply must still explain a blocker and link a saved document. > - This pull request removes repeated procedure text and keeps these visible outcome requirements explicit. > - A document receipt supplies the exact link, and stricter evals check the persisted reply and browser navigation. ## Linked Issues or Issue Description Refs: #14961. Related: #14948 and #15007. **What happened?** Native task envelopes repeat completion procedure text. A reduced envelope needs explicit final-reply requirements. The `write_document` receipt also lacks a canonical document link. **Expected behavior** Keep the completion tools as the source of procedure details. Require one accepted completion result before the final reply. A blocked reply must explain the reason, owner and unblock action. A document reply must contain a working link to the saved document. **Steps to reproduce** 1. Run the native assigned-skill document case and native blocker case. 2. Inspect the run-attributed provider final and its persisted comment. 3. Check the blocker explanation or open the final reply's document link. ## What Changed - Remove repeated completion procedure text from the native task constraints and backend instructions. - Keep explicit blocker and document-link requirements in full task turns. - Return a company/task-scoped `documentHref` from `write_document`. Preserve the link in the idempotent mutation receipt. - Repeat canonical links for this run's current saved revisions in accepted completion feedback. Give blocked providers final-response guidance for the cause, owner and unblock action. - Keep internal document/comment anchors when Markdown issue links load cached issue details. - Add a manual six-cell comparison suite with strict source, build, default-instruction and budget admission. - Capture eighteen shared runnerd RPC projections and six direct OpenCode HTTP projections across start, resume and continuation phases, using scripted local transports and no provider execution. - Apply v3 checks only to the manual instruction comparison; preserve v2 checks for the existing native completion suite. Check the actual persisted blocker reason and exact saved-document link. Click the rendered document link and check the original content marker in the classic document card or the new document tab. - Forward exact OpenCode finishing calls through the controller. Wait for acceptance, keep accepted feedback and concrete rejection text, and reject malformed responses. Preserve ordinary dynamic-tool response handling. - Settle the completion decision and tool response before mapping a racing idle/error/abort event or handling explicit close/interruption. Reject a concurrent finishing call before controller admission. - Add a provider-free regression through real runnerd, the OpenCode proxy and a fake provider. Reject the first completion, accept the corrected report in the same turn, and propose one result. - Keep all original verdicts unchanged. Treat replay under new checks as separate diagnostics. ## Verification - `pnpm -r typecheck` and `pnpm build` pass locally. - Native document-authority tests pass, including company/run authorization and idempotent replay. - Native runtime-context, backend and measurement tests pass. - Final-answer calibration, protocol scoring, source-admission and catalog tests pass. Wrong reasons, absent links and wrong link targets fail. - `pnpm test:e2e:runner:typecheck` passes. Discovery lists exactly six single-attempt local cells with the declared models. - Exported `prepareNativeInstructionPreflight` then `verifyNativeInstructionPreflight` pass on this clean committed source. They build locally and make zero provider calls. - Corrective live confirmation is incomplete. Source 3a7349d passed both Claude and both Codex cases. OpenCode saved the correct document but omitted its final link; its blocker case was canceled before paid execution. Preserve this failure. The e171282 confirmation was stopped during build after fresh review found a completion-settlement race; it executed zero providers. Source c3e0cb303 fixes that race. Two affected OpenCode cases await fresh review and one bounded confirmation; earlier results remain attributed to their original source. - OpenCode proxy parsing, driver, factory and input tests: 81 pass across retained focused runs, including six settlement races. Evaluator/scoring/admission checks: 122 pass. The real proxy regression passes. Fresh local prepare then verify passes with 18 shared and 6 direct scripted captures, fresh SDK/Rust builds and zero providers. - The full local suite recorded two failures: a webhook timeout and a Git-scan load count mismatch. Both files pass in isolation with unchanged assertions/time budgets; preserve the original failure log. Fresh c3e0cb303 CI and review are pending. This PR remains draft. ## Risks - Final-answer wording can vary by provider. The checks cover the declared release-access blocker and saved document fixture, not general answer quality. - A single trial does not establish general equivalence, cause, speed, cost or live resume behavior. - `documentHref` is an additive receipt field. It points to the current saved document, not an immutable historical revision. Replaying an older receipt does not fabricate a new link. - The correction adds four production paths for document receipts, accepted completion feedback and UI navigation, plus four OpenCode controller/proxy paths, beyond the original three instruction paths. Completion rejection must remain repairable; the production-boundary regression covers it. - Preserve the frozen comparison context for live measurement. A merge-tree check against current master is clean. Do not relabel earlier live results as results from a later source tree. ## Model Used - OpenAI Codex, GPT-6 family. The exact serving model ID and context window are unavailable in this session. Capabilities used: reasoning, code editing, shell execution, test authoring and evidence 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 - [ ] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-05 08:10:14 -05:00
3 people	1c07b5903b	feat: Chat leads the left nav, agent work beside chats, and a Combined Inbox + Task List flag (#15100) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work > - The left nav is the main way people move between tasks, the inbox, and Agent Chat > - The nav has separate Inbox and Tasks rows that show overlapping work, and Chat is one row among many > - The side panel beside a chat shows the conversation's own artifacts, not the work the agent did > - People want Chat to be easy to find, and they want one place for their task views > - This pull request moves Chat to the top of Work, shows the agent's tasks and artifacts beside each chat, and adds an experimental flag that folds Inbox into Tasks > - The benefit is a shorter nav and a chat view that shows what the agent is working on. Both changes stay off until an operator enables them ## Linked Issues or Issue Description Refs #14706 (the secondary Agent Chat navigation this change builds on) Refs #14848 (reopen the last visited agent chat) **Subsystem affected** UI navigation (left nav, mobile tab bar), Agent Chat side panel, task list and inbox, and the company artifacts API. **Problem or motivation** Inbox and Tasks are two nav rows for overlapping work. Chat sits in the top group with no clear home. The chat rail lists only agents you already talked to, so you cannot see your other teammates there. The side panel beside a chat shows only the conversation's own artifacts. It does not show the tasks and files the agent made. **Proposed solution** With Agent Chat on, Chat leads the Work section and the rail lists every eligible agent. The chat side panel opens on the agent's tasks as cards, and the agent's artifacts are available from +. A new experimental flag, Combined Inbox + Task List, makes Inbox a set of views inside Tasks. **Alternatives considered** Rebuilding the inbox inside the task list. Instead, `/issues` hosts the existing Inbox component for inbox views and the existing task list for status views, so all inbox behaviour stays the same. **Roadmap alignment** Agent Chat (ROADMAP.md, "Agent Chat (including CEO Chat)"). All changes are behind experimental flags that are off by default. ## What Changed - **Agent Chat nav (streamlined shell):** Chat is the first row of Work, not a top-group row. Workspaces leaves the nav while Agent Chat is on. The mobile tab bar is Home · Chat · + · Tasks · Agents. The legacy shell keeps master's top-group Chat row. - **Chat rail:** `AgentConversationsSidebar` lists every eligible agent. The open chat is first, then conversations by recent activity, then the rest of the roster alphabetically. Terminated agents and agents you left are omitted unless you have history with them. The picker still marks only real conversations as "Open chat". - **Chat side panel:** a new default Tasks tab shows one card per task the agent created, was assigned, commented on, or acted on, newest first. It has the task list's filter popover and a sort control. **+ → Artifacts** shows the agent's artifacts as cards. Cards open in a new tab. Agent Chat off keeps the old Artifacts tab. - **Artifacts API:** `GET /api/companies/:companyId/artifacts` accepts `agentId`. The filter applies to documents, work products, and attachments by the agent each result is attributed to. The shared validator and the UI client carry the new parameter, and the OpenAPI entry picks it up from the shared schema. - **Combined Inbox + Task List flag (`enableCombinedInboxTasks`, off by default):** new card in Settings > Experimental. The Inbox row goes away and its badge moves to Tasks. A Views menu on `/issues` covers Mine, Unread, Blocked, Recent, Everything, All, Active, Backlog, and Done. Bare `/issues` opens the last-used view (default Mine). Links that carry `assignee`, `workspace`, `participantAgentId`, or `q` open All so the filter is kept. `/inbox/*` and `/issues/{all,active,backlog,done,recent}` redirect to the matching view. `/inbox/requests` stays its own page. - **Task detail breadcrumb:** the view key now decides the source, so quick-archive still works after a reload from an inbox view. - **Docs:** `doc/PRODUCT.md` and `doc/SPEC.md` describe the chat rail, the chat side panel, and the new flag. ## Verification - `cd ui && npx vitest run --no-file-parallelism src/components/chat src/components/task-side-panel/TaskSidePanel.test.tsx src/components/AgentConversationsSidebar.test.tsx src/components/Sidebar.test.tsx src/components/SidebarCompanyMenu.test.tsx src/components/Layout.test.tsx src/pages/AgentChats.test.tsx src/pages/InstanceExperimentalSettings.test.tsx src/lib/task-views.test.ts src/lib/issueDetailBreadcrumb.test.ts src/pages/Inbox.test.tsx src/pages/Issues.test.tsx src/App.test.tsx src/App.activity-routing.test.tsx src/components/MobileBottomNav.test.tsx src/components/CommandPalette.test.tsx`: 20 files, 356 tests pass. - `cd server && npx vitest run src/__tests__/company-artifacts-service.test.ts`: 13/13 pass, including the new agent-filter test across all three artifact sources. - The new rail test fails against the unmodified rail. - `pnpm check:token-gates`: all gates clean. - Manual: enable Agent Chat in Settings > Experimental. Open Chat. The rail lists all agents. Open a chat. The side panel shows the agent's tasks. Use **+ → Artifacts** to see the agent's artifacts. Then enable Combined Inbox + Task List. The Inbox row goes away, and Tasks shows a Views menu. - Snapshot baselines are intentionally not updated. See `doc/design/DECISION-SHEET.md`, "Per-change snapshot verification demoted to dormant (Jul 13 2026)". ## Risks - With both flags off, the app behaves like master. The only exception is the API: it accepts a new optional query parameter. - With Agent Chat on, the rail can list many agents in a large company. It uses the agent list the app already loads, and search filters it. - The Tasks panel reads at most 200 recently updated tasks per agent and says so when it reaches the limit. The Artifacts panel reads at most 500 of the agent's artifacts. - Combined Inbox + Task List changes what bare `/issues` opens for people who enable it. Deep links with a task filter still open All. ## Model Used - Claude (Anthropic), model ID `claude-opus-5-5`, through Claude Code with tool use (shell, file edit, test runs). Extended thinking was enabled. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] All Paperclip CI gates are green - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge 🤖 Generated with [Claude Code](https://claude.com/claude-code) --------- Co-authored-by: scotttong <squadbot000@gmail.com> Co-authored-by: Claude Opus 5.5 <noreply@anthropic.com>	2026-10-04 02:13:46 -07:00
DottaandPaperclip	eb049aebf2	feat(skills): let agents update company skills safely (#15049) ... ## Thinking Path > - Paperclip is an open source control plane for AI-agent companies. > - Company skills give agents reusable work instructions. > - Skill Studio can edit skill files and save version history. > - Agents can create a skill, but they do not have a first-class update tool. > - An agent update needs a version check and safe retry behavior to prevent lost edits. > - This pull request adds `update_skill` through the existing company skill file API. > - The change keeps company policy, version history, and audit records in one path. ## Linked Issues or Issue Description **Subsystem affected** Cross-cutting: the server API, shared validation, and runner tool catalog. **Problem or motivation** An agent can create a company skill but cannot update its `SKILL.md` through a first-class tool. An unguarded retry can also create duplicate versions or overwrite a newer edit. **Proposed solution** Add `update_skill` with a required current version ID and a retry key. Route it through the existing skill file API. Reject stale versions and changed-input retries. Save the version and audit event together. **Alternatives considered** A separate write endpoint would duplicate the Skill Studio mutation path and policy checks. This PR reuses that path instead. **Roadmap alignment** This work extends Skills Manager and Skill Studio, which are listed in `ROADMAP.md`. **Additional context** The tool accepts a complete `SKILL.md`, not a partial patch. Callers must read the current version before they edit it. ## What Changed - Add optional version and retry fields to the existing skill file update contract. - Add a guarded API update with a stable retry receipt and attributed audit event. - Add `update_skill` to native and semantic runner tool catalogs, with mode and policy gates. - Add unit, integration, protocol, and semantic-tool regression coverage. - Document agent use and extend the OpenAPI request contract. ## Verification - Focused tests and direct server and runner TypeScript checks passed before this PR. - `git diff --check` passed after the rebase onto `master`. - CI passed on the latest PR head, including the full test matrix, typecheck, and build. Local full typecheck and build stopped because `cargo` is not installed. The local full test run ended without a verdict. - No dedicated end-to-end eval scenario was added or run. The protocol coverage and semantic-tool test cover the new action deterministically. - Reviewers can read a skill version, call `update_skill`, repeat the same key, then try a stale version and a changed-input key. Only the first edit must create a new version. ## Risks - File writes and database transactions must stay in sync when a write fails. The integration tests cover failed writes and retry behavior, but CI must verify them on the PR head. - Existing Skill Studio callers do not send the new optional coordination fields. Their request shape remains valid. ## Model Used - OpenAI Codex CLI assisted with this change. The runner did not expose the exact model ID or context window. The agent used 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; exact model ID and context window were not exposed) - [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; full suite is pending CI) - [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 (53 pass, 4 skip on the latest head) - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-03 12:37:55 -05:00
DottaandPaperclip	b17019e14d	fix(agents): reduce default instructions and qualify stock harnesses (#14948) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Its adapters supply task context and access to Paperclip skills and tools. > - The default hire manual and shared prompts also repeat general work procedures. > - Those procedures overlap with stock provider instructions and the Paperclip skill. > - Existing E2E fixtures supply a QA manual, so they do not qualify the production default. > - This pull request reduces the generic instructions and adds real default-hire coverage. > - The benefit is less competing guidance, with inspectable evidence for preserved skills and task context. ## Linked Issues or Issue Description Refs: #14920. That merged change preserves native Codex base instructions. This PR covers the default manual, shared legacy prompts, operational skill guidance, and the narrowly approved ACP skill-discovery/session-environment repair for measured delivery and credential-persistence failures. **What existing behavior does this improve?** New non-CEO hires without a custom bundle and legacy task/chat startup and continuation prompts. **Current behavior** The shipped default manual contains 602 words. Generic task/chat prompts and ordinary resume deltas repeat work procedures already available through the harness and Paperclip skill. **Proposed behavior** The default manual contains only the eight-word company identity. Shared startup prompts retain identity and connection guidance. Ordinary resume deltas retain current work context without the generic execution contract. **Reason and benefit** Let the stock harness guide general work. Keep Paperclip-specific capabilities and independently test default hires, skills, ordered comments, and chat restart. **Breaking changes** New default hires receive less guidance. Existing saved manuals, explicit custom bundles, CEO templates, and specialized wake contracts retain their behavior. The obsolete includeExecutionContract option remains accepted for source compatibility. ## What Changed - Reduce the default hire manual to one sentence. - Reduce shared task/chat defaults and remove the generic ordinary-resume contract. - Keep connection guidance, auth, skills, custom prompts, and specialized wake context. - Add credential-free instruction-boundary gates and 26 explicit Product E2E cells across eight legacy/native profiles, including two focused Paperclip-storage cases. - Capture public hire receipts before providers run, then grade delivered prompts and independent task/chat outcomes. - Add an early legacy skill API recipe for saving a task document, checking the saved revision receipt and linking the document. Improve stock task/heartbeat skill-selection metadata and show a clickable Markdown UI-link example. Keep native tool completion separate. - Advertise bounded routing descriptions and exact successfully staged SKILL.md paths in legacy ACP Claude; keep full bodies on demand and preserve remote path rebasing. - Remove only the provider environment from copied persisted ACP session records, while loading current run credentials and preserving all other options/conversation state. - Regenerate both capability metadata inventories and reject stale manifests/inventories before provider admission. - Publish the original reduction and focused skill-repair comparisons, preserving all failures, automatic recovery, cost coverage and limitations. ## Verification **Behavioral qualification remains pending.** Original legacy ACP Claude loses the issue document only in the reduced cohort beneath an unchanged credential failure. A source-backed diagnosis finds that neither ordinary assignment reads the staged operational skill, while the runtime persists provider environment in session state. The new common repairs expose skill metadata/path and omit persisted env; strict document and credential guards stay intact. [Inspectable diagnosis and retained hashes](https://github.com/paperclipai/paperclip/blob/9f654db4541d3d002769c988f6e51fc0b08dadbd/doc/plans/2026-10-03-legacy-acp-claude-readiness.md). Current repair head `de0965984ff3edf611ae6d0e7ca5c7d5ae3947bb` incorporates master `569c7203aa24b95440682983ce7940ba1d4247bd` (merged #14961/#15007). All 222 affected adapter tests, adapter-utils/E2E typechecks, and final 96 variant/grader/retry calibrations pass. The frozen historical comparator is `c25697f4260b6f3adfea143c3ae9932e2f42986d`: 8,280 of 8,291 paths identical, exactly two production instruction paths plus nine declared unit expectations differ. The operational skill/discovery/environment repairs, selected model/profile/task/core grader/auth/permissions/retry policy are identical. Both actual launcher prepare→verify admissions pass with zero providers. [Immutable manifest and exact receipts](https://github.com/paperclipai/paperclip/blob/9f654db4541d3d002769c988f6e51fc0b08dadbd/doc/plans/2026-10-03-legacy-acp-claude-evidence/manifest.json). One original legacy ACP Claude cell per variant is authorized, with enforced single campaign attempts, 12-minute deadlines and company/agent 1,000-cent hard stops; every product recovery run/cost is counted. Actual live outcomes are pending. Current normal CI has one failed server shard and failed aggregate verify under diagnosis; other normal gates including typecheck/build/Rust/all eight browser shards pass. Fresh review completed successfully; the valid historical startup/resume masking finding was fixed with per-invocation task/chat checks and strict complete-snapshot capture, calibrated and resolved. Prior heads, failures and campaigns below remain historical evidence, not checks on this repair head. - Prior head `36aa4d81c49a1a8f6f04b1a068fae19aa901955f` is replayed on merged hiring master `862a5758ba0e88a33232c1f1fa645e85c38a3113`. All 52 current-head checks pass with two intentional Storybook skips, including repository typecheck/test/build and the browser shard. Fresh Greptile is 5/5 with zero unresolved review threads. Exact-head stock prerequisites pass 599 assertions (598 TypeScript + 1 Rust), all six gates and retained receipt verification, zero providers/source errors. Fingerprint `a7f5a22d860a88fe20cce213c6d5e0004788f32c8363930729aea4fd740ad16d`. Combined catalog/hiring calibrations pass 67 assertions, E2E typecheck and 26-cell stock discovery pass. Canonical contract/inventory checks and the later issue-derived reference calibration are retained; that reference-only follow-up is not live-qualified by earlier frozen runs. - Prior full repository typecheck/build passed. The complete local Vitest run executed 14,956 tests: 14,870 passed, 83 skipped, three timing failures. All three affected files passed unchanged narrow reruns; original failures remain retained. Current-head CI now passes the full general checks; the original local failures remain retained. - The original 24-pair default-manual/shared-prompt comparison has two new overall classic Claude/OpenCode document-delivery failures plus an additional legacy ACP Claude document loss beneath an unchanged credential-guard failure (not closed by later runs), two newly passing OpenCode ordered cases, seven unchanged failures and 13 unchanged passes. Equal 15/24 totals do not establish behavioral equivalence. [Complete original report](https://github.com/paperclipai/paperclip/blob/875f4c397d9e8c3f12f39dedd59abaf1eaf5236e/doc/plans/2026-10-02-stock-harness-live-comparison.md). - The skill-only repair holds the eight-word manual/shared prompts and merged #14920 fixed. All four matched profile configurations and 203 fixture/behavior files match. Candidate `abd0b628ca642c09a54a4edc56a5227402f6686e` varies only the two skill sources against baseline `bc83fe030234439ac51279502a28803958963e2e`. [Candidate workflow](https://github.com/paperclipai/paperclip/actions/runs/37060885547) and [baseline workflow](https://github.com/paperclipai/paperclip/actions/runs/37060888047) each pass 571 exact-source prerequisites before providers; all eight cells clean up successfully. Failed campaigns publish successfully and remain failed. - Repair pairs: Claude original Fail → Pass; Claude explicit Pass → Pass; both OpenCode cases Fail → Fail. Explicit OpenCode's handoff worsens beneath the unchanged failing UI-link grade: baseline gives a clickable API URL, candidate gives a code-formatted path without an anchor. The request's usable-link wording is narrower in the UI-only oracle. [Complete repair report and safe projection](https://github.com/paperclipai/paperclip/blob/875f4c397d9e8c3f12f39dedd59abaf1eaf5236e/doc/plans/2026-10-02-legacy-document-skill-repair.md). - The subsequent narrow stock metadata/link correction has two matched Pass → Pass cases, zero new machine failures/passes and no pending pairs. Both original-case handoff links remain deficient: candidate uses a wrong PAP prefix, baseline supplies a bare prefix-less slug path; the preserved original oracle only requires a durable document. Both explicit clickable UI-link cases pass revision/content/link grading. All four exact-source 587-check gates, single assignment runs and cleanup pass. This does not establish fix causality because baseline also succeeds. [Candidate workflow](https://github.com/paperclipai/paperclip/actions/runs/37069547401) freezes `fe9dc1e3c518825242ed889ab9c8352986f8c2ed`; [matched baseline](https://github.com/paperclipai/paperclip/actions/runs/37069552374) freezes `0d7ecfa96d72fba79b7f0a25052b42c0686c0488`. This is a skill-only comparison with reduced manuals/shared prompts held constant, not a repeat of the historical-manual comparison. Only original and clarified explicit classic OpenCode cases are selected, two per variant/four expected turns. 8,242 other tracked files and both profile hashes match; protected workflows admit each exact source before credentials. [Complete qualification report](https://github.com/paperclipai/paperclip/blob/74d0d3d945f4c52d0814b5a845ab5bd09f33cd6b/doc/plans/2026-10-02-opencode-skill-routing-link-qualification.md). Candidate original loads Paperclip/reference before saving publicly; baseline original loads it after writing locally, then saves publicly within the same assignment. Reported cost totals are $0.0107824490 candidate / $0.0107909015 baseline, with unmetered runtime. The later reference-only issue-derived link correction is provider-free calibrated and **not live-qualified** by these frozen runs; no further paid runs. - Retained tool calls show the repaired original OpenCode assignment loads only its assigned output skill before writing locally. Operational Paperclip is first loaded during automatic disposition recovery; its early recipe is visible then, but it never saves the missing document. Explicit candidate loads Paperclip and reads the new reference before saving successfully. All nine actual runs are counted. Reported LLM totals are $0.3802537209 baseline and $0.4918990161 candidate; local runtime is unmetered. - Initial setup, packaging, cancelled/missing-cell recovery, callback test and relative-output attempts remain retained. No completed provider failure was rerun. Frozen measurement branches are unchanged by later canonical metadata maintenance. - Run `pnpm test:e2e:runner:stock-harness`, `pnpm test:e2e:runner:unit`, and `pnpm test:e2e:runner:typecheck`. Select `stock-harness` explicitly for paid execution; it is excluded from `--all`. Prior-head integration: `36aa4d81c49a1a8f6f04b1a068fae19aa901955f` replays this PR on merged hiring #14985 (`862a5758ba0e88a33232c1f1fa645e85c38a3113`), preserving the four explicit custom-CEO-bundle checks, minimal generic manual boundary, and both suites. The combined fixture catalog and hiring calibrations pass 67 assertions; exact-head stock prerequisites pass 599 assertions (598 TypeScript + 1 Rust), all six gates and retained-receipt verification, zero providers/source errors, fingerprint `a7f5a22d860a88fe20cce213c6d5e0004788f32c8363930729aea4fd740ad16d`. E2E typecheck and 26-cell stock discovery pass. Fresh current-head CI passes all 52 checks with two intentional skips, and fresh Greptile is 5/5 with zero unresolved review threads. The prior source-plan browser failure is retained: a deterministic process fixture replayed its last `fixture:plan` command on `chat_task_completed`, writing revision 2 with identical body after the approval handoff. This was not paid provider execution. Rebased current-head CI passes the same assertion without an old-head retry or a change to that browser fixture. The merged hiring change was measured separately on immutable matched unions, with this reduced/shared/operational context and native completion guidance held constant. [Complete original two-profile report](https://github.com/paperclipai/paperclip/blob/f0512647656be78e48abd8c22a3078db8bf6bcd2/doc/plans/2026-10-02-hiring-template-live-comparison.md): [candidate](https://github.com/paperclipai/paperclip/actions/runs/37075466208) / [historical baseline](https://github.com/paperclipai/paperclip/actions/runs/37075469463), 705 provider-free prerequisites each. Both pairs are unchanged Fail → Fail on the exact-five count, with six core delivery checks passing all four cells; 28 actual successful runs include eight automatic completion wakes, zero retries, four successful cleanups. Source-read coverage is uncomparable, actual model charges unknown. Separately versioned provider-free accounting remains analytical work; original verdicts are preserved. This does not rerun or qualify the completed default-manual or native campaigns. ## Risks - Legacy ACP Claude's additional delivery loss is not closed by any later matched run and blocks the no-extra-failing-behavior merge criterion. Legacy document delivery may have relied on the prior manual/shared prompts. The early skill repair improves Claude in one trial; the later OpenCode pairs pass in both variants and cannot establish causality or robust recovery. Both original-case links remain deficient beneath the storage-only grade. The later issue-derived reference correction has only provider-free validation. Native finish/block descriptions must not be supplied to legacy agents. - The comparison holds merged native Codex fix #14920 constant; it cannot measure that fix's before/after task performance. - These bounded skill/context/chat workflows do not measure general coding quality. Unrepresented providers remain unqualified. - Saved manuals and old Codex sessions are not automatically migrated. Codex through ACP still has a separate base-instruction follow-up. ## Model Used OpenAI Codex, GPT-6 family as identified by this session. The exact deployment ID and context-window size are not exposed. The assistant used reasoning, repository tools, code execution, and delegated PR/eval work. ## 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 (relevant suites and all three unchanged narrow reruns pass; complete-run timing failures retained in Verification) - [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 on the new repair head (prior-head checks retained above) - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups on the new repair head - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-03 12:32:42 -05:00
DottaandPaperclip	78e0034498	fix(evals): account for hiring completion notifications (#15007) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Product E2E evals check real hiring and delegated task completion. > - The hiring fixture requires three requested CEO turns and two coder executions. > - The server can also wake the CEO when each delegated task completes. > - Two exact-five-run guards rejected these valid completion turns in all four retained cells. > - This pull request validates bounded completion turns in both guards. > - The benefit is accurate workflow grading while all actual runs and coverage failures remain visible. ## Linked Issues or Issue Description Refs: #14985, #14948, #14961. **What happened?** The original hiring comparison reports Codex Fail → Fail and Claude Fail → Fail. Each cell has seven successful runs. The five requested work turns are accompanied by two server task-completion notifications. All six other delivery checks pass. **Expected behavior** Require exactly three distinct user-requested CEO turns and one coder execution for each of two known tasks. Admit at most two strictly attributed server completion turns, including one turn that batches both tasks. Reject unknown, duplicate, failed, retried or extra-work runs. **Steps to reproduce** Inspect the retained four-cell report linked below. Each original result fails `five-successful-turns`. The same exact count was also enforced by the final chat-flow guard. ## What Changed - Add one typed lifecycle helper shared by the hiring scorer and the hiring-only final chat guard. - Validate public run ledgers, company/user/account identity, request attribution, task origins, completion deliveries, timing and replies. - Keep exactly five required work turns; declare seven maximum total turns for cost and timeout planning. - Count all actual runs, including notification runs and unexpected resets. Keep other chat count guards unchanged. - Version the hiring grader as v3 (turn accounting v2) and include the helper and chat guard in its definition digest. - Keep source-read, exact coder-body and all six other delivery checks unchanged. - Add 144 focused helper/scorer/settlement calibrations and separately versioned exact retained-input replay reports. - Retry complete bracketed observations, await both owed callbacks and attributed replies, and refresh the final guard consistently. - Reject unrelated completion writes and failed mutation attempts using exact canonical/native action IDs. Missing identity mapping is uncomparable action coverage. ## Verification - All 977 credential-free E2E support tests pass across 64 files, including 144 focused lifecycle/action/scorer/settlement calibrations. - E2E typecheck, ordinary plugin SDK and Runner TypeScript dependency builds, capability contract/inventory checks and the existing two-cell hiring discovery pass. - [Executable replay report](https://github.com/paperclipai/paperclip/blob/fed1729018cc100f5f4bbfb692777e49009c423b/doc/plans/2026-10-02-hiring-executable-accounting-replay.md) pins current code revision `e4077ade1818d98b9862ae79ee1d49a007dcf9c1`, v3 definition digest, exact source/input hashes and each original/new check. - The stricter replay verifies both Codex variants through both executable guards. ACPX Claude action attribution remains unresolved/uncomparable because provider execution IDs cannot be exactly joined to native request IDs; guards fail closed. No notification writes are observed. All six other outcomes and every original source/template coverage check stay unchanged. Original files and Fail → Fail machine verdicts remain preserved; zero providers are called. - Full attempts remain uncomparable in both profiles. Historical Claude also keeps its six-backtick exact-template mismatch. This grading repair does not prove model-performance equivalence. - The limited sidecar-v1 and initial executable-v2 passes checked notification-created tasks but could miss unrelated document writes. Those assessments remain preserved and do not prove harmless notifications. The stricter v3 replay is separate. - [Original measurement and separate sidecar](https://github.com/paperclipai/paperclip/blob/8eb517ca1497687237163bdef4dfc4d3332ea916/doc/plans/2026-10-02-hiring-template-live-comparison.md) retain 28 actual runs, eight automatic notifications, four successful cleanups and unknown actual model charges. No models are rerun. - The branch is replayed on master `59c07ede7`. Intervening master changes are UI-only; eval source bytes and replay verdicts match. The four-cell provider-free replay was repeated against the reachable code revision. - Initial-head normal CI retained browser failures in agent-run denial feedback and touch-picker scroll position. Those browser paths and imports were unchanged, but their cause was not established. The necessary review-fix head passes both browser checks; no blind rerun was requested. - Local full repository typecheck/test/build were not repeated. Exact-head normal CI passes the required repository gates, including typecheck, tests, build and browser shards. Fresh Greptile review completed on `fed1729018cc100f5f4bbfb692777e49009c423b` with 5/5 and zero unresolved threads. An independent rerun of the 144 focused helper/scorer/settlement tests passes on the unchanged head. **Merge readiness:** This PR repairs the evaluator. Its positive and negative calibrations pass, both guards reject missing action attribution, current-head CI and review pass, and there are no merge conflicts. The retained ACPX cells remain uncomparable because their action IDs cannot be joined. That coverage limit remains a separate follow-up; it does not require relaxing this grader or changing the old results. No model calls, production instructions, carrier changes, or historical regrades are part of this readiness update. ## Risks - Missing or inconsistent public lifecycle evidence fails the bounded helper. The focused calibrations reject plausible false positives and malformed observations. Unmatched action IDs fail closed and are reported as uncomparable rather than a model task regression. - Source-read evidence remains incomplete. This PR does not change provider event carriers or relax the coverage oracle. - The versioned count check differs from original v1 results. Reports retain both versions and exact input hashes. ## Model Used OpenAI Codex, GPT-6 family as identified by this session. The exact deployment ID and context-window size are not exposed. The assistant used reasoning, repository tools, code execution and delegated calibration work. ## 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 normal CI gates are green (exact head `fed1729018cc100f5f4bbfb692777e49009c423b`; fresh review tracked separately below) - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups (completed exact-head review; zero unresolved threads) - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-03 06:42:51 -05:00
DottaandPaperclip	ffe5e9e2a8	fix: retain execution evidence for Retry and saved input (#15033) ... ## Thinking Path > - Paperclip lets people steer and recover AI-agent conversations. > - Recovery eligibility depends on retained cancellation receipts. > - Run presentation intentionally omits result JSON on SQL_ASCII databases and reduces oversized output. > - Retry and the saved-input sweep mistakenly used that presentation read for admission. > - The banner could offer Retry while the endpoint rejected the same stopped run. > - Read the narrow execution-evidence fields for internal admission and keep normal presentation unchanged. ## Linked Issues or Issue Description Refs #15024 and #15015. Searched existing recovery and redaction PRs; no duplicate fix found. **What happened?** On a SQL_ASCII instance, a verified pre-dispatch review-wait cancellation offers Retry in the recovery notice. Clicking it returns an eligibility conflict, and a saved user message remains deferred. The notice reads the retained database receipt, but the endpoint and sweep read a presentation projection where `resultJson` is null. **Expected behavior** Retry and saved user input use the recorded execution evidence and ordinary admission gates, independent of presentation redaction. Public run reads retain their existing encoding and output-size protections. **Steps to reproduce** 1. Record a cancelled, unclaimed review-wait continuation and its recovery hold. 2. Use the SQL_ASCII presentation projection, where run result JSON is omitted. 3. Click Retry or save a new user message and let the recovery sweep inspect it. 4. Verify a fresh turn starts once, with no replay of consumed input. **Paperclip version or commit** Reproduced on `9ae3d8db3`. **Deployment mode** Authenticated private self-hosted server with a SQL_ASCII database. ## What Changed - Add an explicit internal read of cancellation, startup, review-wait, tool-inventory, and Stop evidence; omit provider diagnostics. - Use that read in the Retry route, wakeup validation, and saved-input continuation checks. - Preserve the distinction between absent result JSON and an unrecognized stored result. - Cover the reproduced SQL_ASCII Retry and saved-input failures, retained public redaction, native Stop behavior, and excluded provider output. - Document the presentation and admission distinction. ## Verification - Red: four selected assertions fail before the fix, including the SQL_ASCII eligibility conflict and saved input remaining deferred. - Targeted green regressions and existing native Stop cases pass. - Workspace `pnpm -r typecheck` and `pnpm build` pass. - All 626 affected recovery, continuation, and route tests pass, including 22 focused admission and native Stop cases. Current-head CI has 54 passing gates and 2 skipped optional Storybook checks. Greptile reviewed `9b94af91e295a4e8007dfc6bff6a8d532d945e36` at 5/5 with no findings or open review threads. No complete local monolithic pass is claimed; the complete suite runs in sharded CI. ## Risks Admission still checks recorded process and controller ownership, provider events, cleanup, company scope, user authority, pending decisions, and task holds. The evidence projection must retain every field used by these eligibility predicates; existing native Stop cases guard against dropping its acknowledgement receipt. No schema, dependency, UI, or public response change. ## Model Used OpenAI Codex, an agent based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Used reasoning, repository tools, code execution, and browser inspection. ## 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>	2026-10-03 01:38:57 -05:00
DottaandPaperclip	9ae3d8db3d	fix: keep pre-dispatch review waits out of execution recovery (#15024) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The queued-run gate can cancel a continuation that must wait for review. > - This cancellation happens before execution authority or a provider starts. > - Recovery currently treats the gate receipt as unknown provider execution. > - That mistake blocks a conversation after a successful reply and hides Retry. > - This pull request recognizes only the recorded, unclaimed review-wait state. > - Review waits keep their normal disposition path, and new user input can recover older false holds. ## Linked Issues or Issue Description Refs #15015, #15020, and #15022. Related #11614 narrows the review posture that causes cancellation; this change corrects recovery after a valid cancellation. **What happened?** After a successful agent reply, the queued-run gate cancelled an automatic continuation with `issue_continuation_waiting_on_review`. The gate retained `timeoutSource: stale_queued_run_gate` and a matching stop reason. No execution authority or provider started. Recovery still created an unknown-action hold, moved the task to Blocked, and hid Retry. **Expected behavior** Use normal review-wait disposition repair for this recorded state. Allow Retry, a new user message, or saved undelivered input to recover an older false hold after ordinary admission checks pass. Preserve the cancelled run and do not replay its input. **Steps to reproduce** 1. Finish an agent turn on an open task that has a real review target. 2. Let the automatic continuation reach the queued-run review gate. 3. Refresh after the cancelled run is checked by recovery. 4. Confirm a review wait is handled as a wait rather than unknown provider work. 5. Reproduce an older false hold for the same receipt, then request Retry or send a new message. 6. Confirm only one fresh turn starts, and contradictory execution or cleanup evidence retains the hold. **Paperclip version or commit** Reproduced on `215586d127`. **Deployment mode** Authenticated private self-hosted server, built from source. ## What Changed - Recognize the exact review-wait dispatch receipt only while all execution claims remain unset. - Exempt recovery only after checking retained launch and provider events, coordinators, and environment cleanup. Keep the synchronous classifier conservative without that database proof. - Apply the verified classification to automatic recovery, heartbeat retries, and stranded-queue release, so saved user input starts once. - Reuse guarded startup admission for Retry, new input, and saved input on older false holds. - Show a precise review-wait notice and keep continuation guidance consistent with Retry availability. - Verify provider events, launch events, coordinators, and cleanup before admission. - Add five initial red regressions, three additional red recovery evidence regressions, concurrent saved-input coverage, and negative evidence checks. - Document the review-wait contract. ## Verification - Red: five regressions fail on unchanged master. Existing review-wait and contradictory-evidence cases still pass. - Workspace `pnpm -r typecheck` passes. - All 752 affected recovery, continuation, queue, classification, and retry-scheduling tests pass. - Three additional review regressions failed before the database proof was added; all 12 focused review-wait cases then pass. - Workspace `pnpm build` passes. - Saved-input promotion and recovery notice regressions fail before their fixes and pass afterward. - Current-head CI has 54 passing checks and 2 skipped optional Storybook checks. Greptile reviewed `0bf1437110e1617ae4544afa41f4319eac763dba` at 5/5 with zero open findings. The complete suite runs in sharded CI. No complete local monolithic pass is claimed; an earlier long run was stopped, and its unrelated failing case passed in isolation. ## Risks The error code alone cannot establish that no provider started. This exception also requires the server gate receipt, matching stop reason, and null execution authority fields. User admission separately checks coordinator, launch and provider events, environment cleanup, pending decisions, ownership, task holds, budget, and active execution. No migration or dependency change. ## Model Used OpenAI Codex, an agent based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Used reasoning, repository tools, code execution, and browser inspection. ## 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>	2026-10-03 01:00:42 -05:00
DottaandPaperclip	215586d127	fix: settle interrupted preparation with a retained cancellation fence (#15022) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A cancelled preparation must preserve history while allowing a new user turn. > - Older preparation can retain its cancellation fence but omit the unwind marker. > - The controller from that older boot is gone, its lease expired, and no provider started. > - Use the same narrow preparation proof for this retained-fence state. > - The benefit is working Retry and saved-message recovery after interrupted startup. ## Linked Issues or Issue Description Refs #15020 and #15015. Related #13293 addresses post-launch recovery. **What happened?** A preparation interrupted before native selection retained `startupCancellation.beforeNativeSelection: true` but no preparation-settled marker. Its owner expired across restart and it had no environment leases. The missing-receipt compatibility rule did not recognize the retained fence, leaving Retry absent and saved messages deferred. **Expected behavior** Admit one new user turn when the preparation evidence agrees, the old controller expired, and cleanup is complete. Preserve previous results and do not replay the cancelled input. **Steps to reproduce** 1. Cancel Paperclip Runner preparation before runtime selection. 2. Retain the cancellation fence without an unwind marker or environment leases. 3. Restart after its old controller lease expires. 4. Send a new message, select Retry, or allow the saved-message worker to reconsider new input. 5. Confirm one successor receives only undelivered input. Repeat with live ownership, invocation evidence, or pending cleanup and confirm execution stays held. **Paperclip version or commit** Reproduced on `94f6f3eb4`. **Deployment mode** Authenticated private self-hosted server, built from source. ## What Changed - Accept a retained before-selection cancellation fence in the expired historical preparation proof. - Retain runtime, stage, adapter, ownership expiry, process, event, and cleanup requirements. - Extend Retry, fresh-message, partial-queue, concurrent-worker, and contradictory-evidence tests to both receipt states. - Document recovery when the preparation-settled marker was not retained. ## Verification - Red: five exact-state regressions fail on the parent commit. - Green: all 203 continuation tests pass, including the five new regressions and 13 additional negative evidence cases. Process recovery and queued-comment routes add 413 passing tests. - A read-only candidate service check against the retained live run returns Retry eligibility without changing any task state. - An unrelated containment assertion failed once in CI. All 85 tests in that route suite pass locally and the CI shard passes on rerun. - Workspace `pnpm -r typecheck` and `pnpm build` pass. - Final head `48240de3d`: all 54 CI checks pass, two optional Storybook checks skip, and Greptile is 5/5 with zero unresolved threads. The complete local monolithic suite is covered by sharded CI; the earlier local run was stopped after a workspace case failed, and that case passed in isolation. - After merge, deploy the exact merged commit and verify an actual agent reply through the task composer. ## Risks A cancellation receipt alone must not certify that provider execution stopped. This path still requires unresolved preparation, no native identity or coordinator, an expired controller from another boot, no launch or provider evidence, and completed environment cleanup. Current-boot preparation remains held until it settles. No schema or dependency change. ## Model Used OpenAI Codex, an agent based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Used reasoning, repository tools, code execution, and browser inspection. ## 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>	2026-10-03 00:04:21 -05:00
DottaandPaperclip	94f6f3eb47	fix: recover historical interrupted native preparation (#15020) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A task conversation must accept new user input after an interrupted startup. > - Native startup begins with a legacy preparation row before runtime selection. > - Older builds did not retain the startup cancellation receipt on that row. > - The immutable adapter claim and expired controller can still prove that no provider started. > - This pull request uses that narrow proof for explicit Retry and saved user input. > - The benefit is a conversation that recovers after an upgrade without repeating old work. ## Linked Issues or Issue Description Refs #15015. Related #13293 covers retained process evidence after provider startup. This change covers interrupted preparation before native runtime selection. **What happened?** After an upgrade, a task stopped during native preparation can still show automatic recovery stopped. A new user message saves but does not start. Retry is absent because the historical run has no cancellation receipt. **Expected behavior** Offer Retry and admit new user input when immutable run evidence proves that no provider started and cleanup is complete. Preserve incomplete or contradictory evidence as a recovery hold. **Steps to reproduce** 1. Retain a cancelled run with the Paperclip Runner adapter claim, an unresolved runtime, and the preparing stage. 2. Keep its old controller boot ID and expired lease. Retain no native identity, coordinator, result, process identity, or provider events. 3. Upgrade from a build that did not save the startup cancellation receipt. 4. Send a new user message or select Retry. Confirm that one fresh turn starts. 5. Repeat with an active controller, a provider launch, or unfinished cleanup. Confirm that execution stays held. **Paperclip version or commit** Reproduced on `cc67d4e1d` with a historical interrupted preparation row. **Deployment mode** Authenticated private self-hosted server, built from source. ## What Changed - Recognize historical interrupted native preparation from immutable run evidence and an expired controller from another server boot. - Reject adapter invocation evidence in the startup proof. Keep process and environment cleanup checks. - Apply the same proof to saved user messages in the bounded recovery worker. - Recheck saved-message eligibility under the existing task and run locks. Keep normal ownership, decision, pause, and budget gates. - Add regression tests for Retry, a new message, concurrent saved-message recovery, and contradictory evidence. Document the compatibility rule. ## Verification - Red: Retry, new-message recovery, and saved-message recovery fail on the parent commit. - Green: 185 continuation tests, 335 process-recovery tests, and 78 queued-comment route tests pass. Two additional red regressions cover partially delivered saved queues and pass after the admission fix. - Workspace `pnpm -r typecheck` and `pnpm build` pass. - Final head `48db53f4f`: all 54 CI checks pass; two optional Storybook checks skip. Greptile is 5/5 with zero unresolved threads. - The local monolithic `pnpm test:run` was stopped after CI passed. One unrelated workspace case failed in that long run; all three workspace reconciliation cases pass in isolation. No complete local monolithic pass is claimed. - After merge, deploy the exact merged commit and verify recovery through the normal task composer. ## Risks Historical compatibility could grant a new turn without enough startup evidence. The proof requires an immutable native adapter claim, an unresolved preparing stage, no result or native identity, an expired controller from another server boot, and no invocation or provider evidence. Environment cleanup remains mandatory. The fix does not replay old input or change historical run results. No schema or dependency change. ## Model Used OpenAI Codex, an agent based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Used reasoning, repository tools, code execution, and browser inspection. ## 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>	2026-10-02 23:08:05 -05:00
DottaandPaperclip	cc67d4e1d8	fix: preserve steering and recover stopped task conversations (#15015) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A task conversation must let a user guide a running agent and resume stopped work. > - The active run owns its input protocol, even when the user changes the next model or effort. > - Queue delivery waits for a provider receipt, which must be able to persist during the request. > - A stopped startup also needs a clear user action that passes normal task admission. > - This pull request fixes steering delivery, makes queue actions immediate, and restores explicit continuation. > - The benefit is a responsive conversation that can recover without losing saved input. ## Linked Issues or Issue Description **What happened?** A queued message could change from Steer to Interrupt while a native run prepared. A steer request could wait on its own database lock and fail to deliver. A stopped startup could then leave the conversation without a working Retry or message continuation. Interrupt also waited for the server and showed a toast. **Expected behavior** The active run keeps its input protocol. Steer delivers input to that run. Steer and Interrupt clear the submitted queue rows and show the input in the conversation immediately. Failed delivery restores the latest queue with an inline error. An eligible stopped run offers Retry, and authenticated user input can start a fresh turn through normal task admission. **Steps to reproduce** 1. Start a task with a native Paperclip Runner. 2. Change the selected model or effort while that run prepares. 3. Queue a message and press Steer. 4. Observe the provider receipt and queue state during the request. 5. Stop a startup before its provider process begins, then try Retry or send a new message. 6. Repeat queued delivery with a legacy runner and press Interrupt. **Paperclip version or commit** Reproduced on the parent of this branch, `59c07ede7`. **Deployment mode** Authenticated private deployment. The fixes also cover local task conversations. Related work: Refs #12834, Refs #13354, Refs #13275. The open refactor in #13160 moves the same queue route; it does not fix the receipt lock or stopped-run continuation addressed here. ## What Changed - Select queue behavior from the active run's immutable dispatch and runtime resolution. - Leave the run row unlocked during provider acknowledgement, then lock and read it before merging the receipt. - Retain queued input if the target run stops during that wait. Keep inline delivery errors visible after empty queue updates. - Permit exact Retry and authenticated continuation after verified native startup cancellation. Preserve pause, approval, budget, ownership, and process-stop gates. - Carry undelivered native queue input into a fresh turn once the old execution is confirmed stopped. - Show Steer and Interrupt input in the conversation and clear submitted composer rows immediately. Restore the latest queue inline on failure. Remove delivery toasts. - Keep optimistic delivery stable across stale polls, empty queues, and paginated history. Preserve classic Interrupt error handling. - Document recovery and optimistic delivery behavior. Add regression tests across server, shared queue projection, and UI boundaries. ## Verification - Red-green regression tests reproduced the queue protocol, receipt lock, stopped-startup continuation, and optimistic delivery failures. - The focused server route, continuation, queue, and runner boundary suites passed during implementation. - The queue-route suite passes with 78 tests. The three complete conversation UI suites pass with 347 tests. - UI typecheck, production build, and `pnpm check:token-gates` pass. - Workspace `pnpm -r typecheck` and `pnpm build` pass. The local monolithic `pnpm test:run` is still running; remote CI verifies the complete suite on the latest commit. - All CI gates pass on `bd9031ad56abfcde13d13a13488c1b9217c2fd3a`, including the full test shards, runner verification, browser E2E, typecheck, release registry, and canary dry run. - Greptile reports 5/5 for that commit. Both review threads are resolved. ## Risks This changes queue display and explicit continuation admission. The UI must restore rejected delivery without losing other-session edits. The server must preserve concurrent provider result updates and must not resume a process whose stop is uncertain. Focused tests cover these boundaries. This change has no database migration. ## Model Used OpenAI Codex, an agent based on GPT-6. The exact runtime model ID and context window are not exposed in this session. Used reasoning, repository tools, code execution, and browser inspection. ## 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>	2026-10-02 22:07:14 -05:00
Devin FoleyandPaperclip	1815474597	fix: report pending execution phase at Stop timeout (#14990) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The server owns each adapter execution and waits for it to settle after Stop. > - A Stop timeout reports that termination remains unverified. > - Existing phase timings arrive only after their work completes, so a stalled await has no timing. > - This pull request samples the pending phase when the Stop timer expires. > - Operators can identify the pending operation without treating diagnostics as stop proof. ## Linked Issues or Issue Description **What existing behavior does this improve?** The opt-in Sentry context for an unconfirmed adapter Stop timeout. **Current behavior** The timeout includes execution identity but no pending phase. A session close, instruction collection, workspace restore, or diagnostic write can remain pending without producing its completion timing. **Proposed behavior** Add a closed-list phase and elapsed milliseconds from the exact live execution control. Sample them when the timeout fires. Report `unknown` and a null age when the control or attribution is unavailable. **Reason and benefit** The next timeout can identify which operation is still pending. It does not require task text, paths, provider output, or additional database writes. **Breaking changes** No API or execution behavior change. The existing opt-in error context gains two fields. Related public work: #14639 added Stop identity diagnostics; #14866 and #14945 cover instruction cleanup and teardown outcomes. This change adds pending attribution to those paths. The native Stop work in #14802 remains separate. ## What Changed - Add a bounded tracker per execution control. Token scopes support nested and overlapping awaits. A late release cannot clear a newer scope. - Track adapter execution, ACP cancellation and settlement, diagnostic writes, and host cleanup. Keep a coarse host scope until the executor finishes. - Sample only the matching current control and settlement promise at timeout. Freeze the sanitized result. Use a monotonic clock and cap elapsed time at one day. - Test stalled operations, repeated Stop calls, stale and wrong-run controls, callback failures, scope bounds, and the real Sentry SDK context. ## Verification - `pnpm -r typecheck` passed. - `pnpm build` passed. - Six focused suites passed: 79 tests. They cover pending scopes, Stop control ownership, real ACP settlement stalls, and Sentry context isolation. - The real Sentry SDK contract ran with the audited optional peer `@sentry/node@10.71.0` installed outside the workspace. Valid phase and elapsed values were exported; arbitrary labels and nonfinite elapsed values were rejected. - An independent agent reviewed the production diff and ran the focused tests without blockers. - `git diff --check` and a redacted Gitleaks scan passed. The local full `pnpm test:run` was stopped during its large serial server batch to avoid duplicating the sharded CI suite. No complete local broad-suite pass is claimed. - All CI gates passed on `c223237b58aa8d479d1c66d7d399de30270bac4f`: 54 successful checks and two expected Storybook skips. This includes the full sharded test suite, typecheck, build, real Sentry SDK isolation, browser tests, canary dry run, and the security scan after the PR became ready for review. - Greptile scored the same commit 5/5 with no actionable findings or unresolved review threads. ## Risks This is diagnostic instrumentation. Cancellation, deadlines, teardown order, termination proof, and file recovery proof remain unchanged. Unsupported or uninstrumented work uses a coarse phase. Tracker overflow fails closed to `unknown`. The tracker emits no new run-log or Telemetry event. The existing Sentry opt-in gate remains in place. ## Model Used OpenAI Codex, GPT-6. The agent used code inspection, local command execution, automated tests, and an independent agent review. The runtime did not expose a more specific model 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>	2026-10-02 16:55:38 -07:00
Devin FoleyandPaperclip	4abff286c2	fix: retain resolved ACP execution timeout metadata (#14986) ... ## Thinking Path > - Paperclip manages agent work through adapters and heartbeat runs. > - ACP adapters resolve a timeout for the selected execution target. > - An untouched sandbox timeout uses a four-hour default. > - Heartbeat finalization rebuilt the metadata from the stored zero. > - This made a timed-out sandbox run report an effective timeout of zero. > - This change retains the adapter's resolved policy for accurate run diagnostics. ## Linked Issues or Issue Description **What happened?** ACP sandbox runs with `timeoutSec: 0` use the four-hour default. Their terminal metadata reports `effectiveTimeoutSec: 0` and `timeoutSource: config` because heartbeat finalization only reads the stored agent configuration. **Expected behavior** The result must report the policy the adapter used: `14400` and `sandbox_default`. Explicit limits, fractional limits, explicit unlimited overrides, and local defaults must keep their resolved values. **Steps to reproduce** Run an ACP adapter on a sandbox target with `timeoutSec: 0`. Compare the start log's four-hour policy with the terminal result's effective timeout. The new tests exercise the adapter result and the heartbeat metadata merge without waiting four hours. **Paperclip version or commit** Reproduced from `6eaf218924f0a89faf1c02eb0d6877a6c5c8a2cb`. **Deployment mode** Self-hosted server with an ACP sandbox execution target. Searched open timeout and metadata issues and PRs. Related #14804 exposes timeout configuration in forms; #14496 proposes a default policy; #14833 addresses CLI session retention. This PR changes only ACP result metadata. ## What Changed - Retain the resolved timeout in the ACP result after settlement. - Use validated adapter resolution when merging terminal timeout metadata. Preserve config fallbacks for older adapters and the HTTP millisecond policy. - Test sandbox defaults, explicit and fractional limits, explicit unlimited overrides, local defaults, malformed metadata, and unchanged cancellation fields. - Document the result fields and their meaning. ## Verification - `pnpm -r typecheck` passed. - `pnpm build` passed. - Stop metadata tests: 23 passed. - Reporter and diagnostic suites: 84 passed; two real-Sentry-SDK tests skipped because the optional SDK is not installed. - ACP engine suite: all 206 tests passed with a deterministic local `gemini --version` shim. Ambient host CLI probes made the existing Gemini session-resume fixture intermittent (one assertion failure in each of two broad runs); an isolated 15-case rerun and the clean-base 206-test suite also passed. No assertion or timeout was changed. - `pnpm test:run` is in progress. This PR does not claim a complete local suite pass. - Independent review found no blocking issues. The tests cover real adapter emission and the real metadata merge separately. ## Risks Low runtime risk: this changes result diagnostics. It does not change timeout values, cancellation acknowledgement, cleanup, checkpoint safety, retries, or provider operations. It does not fix the cause of a quiet or long-running tool. Older stored results are not rewritten. The new source values apply only when an adapter returns a valid resolution. ## Model Used OpenAI GPT-6 with reasoning, repository inspection, code editing, and test execution. The deployment-specific 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 - [ ] 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>	2026-10-02 16:09:36 -07:00
DottaandPaperclip	862a5758ba	fix(agents): reduce hiring templates to role descriptions (#14985) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - New agents receive role instructions from onboarding, the hiring skill, or a team package. > - These sources repeat harness procedures and impose generic work policies. > - They can crowd out the task and the harness instructions. > - This pull request reduces those sources to short role descriptions. > - It preserves configuration, skills, authentication, reporting lines, and approval controls. > - The benefit is less repeated instruction text with explicit coverage for the default hiring path. ## Linked Issues or Issue Description Refs #3307. The CEO template can impose a fixed delegation route instead of letting the agent choose how to fulfill the request. This change removes that route. It does not implement autonomous goal selection. Related work: #14920 preserves stock Codex base instructions. #14948 reduces the generic manual and shared runtime prompts. #14961 improves native completion-tool descriptions. This PR is separate from those changes. ## What Changed - Select only the short CEO `AGENTS.md` for new default CEO bundles. Keep the three former companion files as compatibility assets. - Reduce the first-agent chief-of-staff prompt and coder, QA, UX, and security role examples. - Reduce seven bundled team role bodies. Preserve their role, reporting, and skill metadata. Regenerate the catalog. - Make hiring examples optional. Replace the long generic role manual with short role drafting guidance. Preserve explicit requester instructions. - Add configuration and import coverage for native and legacy managed bundles, custom instructions, first-agent rendering, and catalog contents. - Add an explicit-only hiring eval that starts from the production CEO default and checks one coder hire, independently computed JSON output, saved instructions, and worker reuse. - Include the full prompt comparison and a separate three-request drafting simulation. Neither is a live provider comparison. Prompt differences: [before and after](doc/plans/2026-10-02-hiring-template-prompt-diff.md). The CEO default falls from 1,897 to 20 words. The coder example falls from 652 to 18 words. Word counts describe instruction size, not outcome quality or billing. ## Verification - PASS: 99 focused server tests and eight shipped-catalog tests. - PASS: catalog generation and validation for four shipped teams. - PASS: hiring skill validation. - PASS: `pnpm -r typecheck`. - PASS: `pnpm build`. - INCOMPLETE: the full local `pnpm test:run` was stopped before rebase. Its original log is retained. This is not a completed full-suite pass. The full current-head GitHub CI workflow passed: https://github.com/paperclipai/paperclip/actions/runs/37073372419. - PASS: `pnpm test:e2e:runner:typecheck` and `pnpm test:e2e:runner:unit` (63 files / 843 tests). - PASS: discovery for the two new hiring cells, 50 existing everyday cells, and the full 438-cell catalog. - PASS after rebase: 99 server tests, 11 catalog tests, 62 selected E2E support tests, and the E2E typecheck. - PASS: all current-head PR checks at `57dcee147ed0b2d2e3cc657cd9e50fb16bf9ec25`: 51 successful check runs, two intentional Storybook skips, and successful Snyk status. Fresh Greptile is 5/5 with zero unresolved threads. - PENDING follow-up: matched live hiring runs on frozen integration refs. No live outcome-quality or non-regression result is claimed from the configuration checks or this merge. The new suite has two local native cells: Codex and ACPX Claude. It expects five provider turns per cell. It compares source-derived bundles, so the historical long templates remain admissible. Missing successful source-read receipts make a pair uncomparable. They do not establish a behavior regression or equivalence. ## Risks - New default roles have fewer prescribed procedures. Live checks must determine whether a removed instruction was needed for an outcome. - Existing custom and saved bundles keep their contents. The retained companion assets avoid a source-file compatibility break. - Specialized Summarizer, Reflection Coach, and Wiki Maintainer prompts remain unchanged. Their product contracts need separate review. - The generic non-CEO fallback reduction is in #14948. This PR alone does not provide its eight-word fallback. - Configuration tests and drafting simulations do not establish live outcome quality. QA, UX, security, and chief-of-staff hiring behavior remain outside the new two-cell comparison. > 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 Codex, based on GPT-6, with reasoning, code editing, shell tools, and delegated verification. The runtime does not expose the exact deployment 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 #` / `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>	2026-10-02 17:56:42 -05:00
DottaandPaperclip	6eaf218924	chore: grant Storybook publishing access through CODEOWNERS (#14984) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Storybook previews help contributors review the board UI. > - The publishing workflow uses the default branch CODEOWNERS file to authorize users. > - Tonio and Scott need access to this workflow. > - This pull request adds both accounts as release documentation owners. > - The existing workflow can then authorize both accounts without a separate user list in code. ## Linked Issues or Issue Description **What existing behavior does this improve?** Access to the Storybook build and publishing workflow. **Subsystem affected** Repository ownership and GitHub Actions authorization. **Current behavior** The workflow reads individual accounts from every CODEOWNERS rule on the default branch. Neither `tonio-alucema` nor `scotttong` is in that file. Both accounts already have repository admin access, but the workflow authorization check denies them. **Proposed behavior** Add both accounts to the `doc/RELEASING.md` ownership rule. After merge, both can start and rerun Storybook publishing workflows. A configured `storybook-deploy` environment reviewer must still approve deployment. CODEOWNERS membership does not add an account to the environment reviewer settings. **Reason and benefit** Contributors can publish UI previews through the same CODEOWNERS policy as other maintainers. The authorization code has no account-specific exceptions. **Breaking changes** None. The existing authorization and deployment approval checks remain in place. **Additional context** Related changes: #13226 added the publishing workflow. #13231 added stable branch bookmarks. A search found no open PR for these permission changes. ## What Changed - Add `@tonio-alucema` and `@scotttong` only to the release documentation ownership rule. - Test that accounts listed for release documentation can start and rerun publishing workflows. - Test that removing an account from CODEOWNERS removes its publishing access. - Explain how CODEOWNERS entries and environment reviewers affect publishing access. ## Verification - `node --test scripts/__tests__/storybook-deploy.test.mjs`: all 25 tests pass. - `actionlint .github/workflows/storybook-deploy.yml .github/workflows/storybook-visual.yml`: passes. - `git diff --check`: passes. - GitHub API checks confirm that both accounts have repository admin access. The deployment environment requires an existing reviewer and disables administrator bypass. - Repository-wide typecheck, tests, and build were attempted. They cannot complete in this worktree because workspace dependencies are not installed. Typecheck cannot find Node type definitions. Tests and build cannot load the workspace `tsx` package. - After merge, run `Storybook Deploy` from `master`, select a source branch, obtain environment approval, and check the published URLs in the run summary. ## Risks Both accounts gain permission to start and rerun Storybook publishing. Deployment still needs approval from a configured environment reviewer. No AWS permissions or live GitHub settings change in this PR. ## Model Used OpenAI GPT-6 through Codex. The exact backend model ID and context window are not exposed in this session. The agent used reasoning, repository inspection, code editing, shell execution, and GitHub API 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 - [ ] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 17:34:37 -05:00
Devin FoleyandPaperclip	22cea6b2e6	fix: bound sandbox bridge waits and flag silent runs sooner (#14979) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Sandbox agents exchange input and output through bridge control commands. > - A provider can stop responding to a command even when it receives a timeout. > - These small commands can inherit a four-hour agent lifetime and block input or teardown. > - The board also calls a silent run healthy for the first hour. > - This pull request bounds bridge control waits and surfaces silence sooner. ## Linked Issues or Issue Description **What happened?** A sandbox run can remain active when a bridge control command never returns. The shared helper passes a timeout to the provider but does not enforce it on the host. It also accepts the agent's hours-long timeout. Output silence remains `ok` for an hour and becomes `critical` only after four hours. **Expected behavior** Bound short bridge operations even if the provider never settles. Report failed input delivery through the existing shutdown path. Warn after five silent minutes and escalate after fifteen. Keep normal agent command limits and require verified termination before releasing execution ownership. **Steps to reproduce** 1. Use a sandbox runner whose bridge read or input-upload promise never settles. 2. Set its configured timeout to four hours. 3. Observe that the old queue client never returns or rejects. 4. Inspect a running task with 35 minutes of output silence. The old summary still reports `ok`. **Paperclip version or commit** Base commit `d6d88b9de2`. **Deployment mode** Self-hosted server with sandbox execution. **Agent adapter(s) involved** Shared command-managed sandbox bridge, including Codex ACP sessions. The informational silence thresholds apply to active runs across adapters. Related: #14889 recovers stalled Daytona output streams; #14485 retries explicit gateway failures during input delivery. This change bounds short control operations whose provider promises never settle. It does not add tool replay or automatic cancellation for output silence. #6297 proposes configurable per-agent silence thresholds; this patch only changes the existing defaults. ## What Changed - Enforce at most 30 seconds per bridge control shell command on the host and provider, including callback startup and shutdown, process-session launch, and payload setup. Preserve shorter configured deadlines and launch environments. - Keep the long-lived agent command outside this deadline. Use a fixed timeout diagnostic without command payloads. - Surface suspicious output silence after five minutes and critical silence after fifteen minutes. - Decouple the shared-workspace holder cutoff from warning thresholds and preserve its existing one-hour value. - Add regressions for hung reads, a late upload response, failed input delivery, exact warning boundaries, and fresh output clearing warnings. - Update the adapter guide and execution contract. ## Verification - The three new queue-client regressions fail on the unchanged base and pass with this patch. - Final callback bridge and sandbox session suites: 214 passed. These cover hung reads, writes, startup, shutdown, process-session launch, payload setup, and the separate long-running agent limit. - Stdin ordering and shutdown suite: 56 passed after the lifecycle change. - Daytona and watchdog coverage passed in the earlier focused runs. Across the focused suites, 602 distinct tests pass. - `pnpm -r typecheck` and `pnpm build`: passed. Server and adapter typecheck/build also passed after their respective follow-up changes. - `pnpm test:run`: attempted and stopped after known local failures. Four chat/email cases used an external ancestor skill path, three skill-cache cases failed on macOS, and one wakeup case timed out. The wakeup case passes alone (1 passed, 27 skipped). This run spanned the workspace-cutoff follow-up and also failed its new holder case; a fresh final-head workspace suite passes all 19 tests. The interrupted run is not a full local-suite pass or final-head verification. - A filesystem queue-drain test failed once during the lifecycle rerun and passed on the complete two-suite rerun. It uses the filesystem client, outside the changed command-runner path. - Complete CI on `ff2212c235`: 53 successful checks and two expected skips, including the full test suite and canary packaging dry run. No failed or pending checks. - Greptile reviewed `ff2212c235` at 5/5. All review findings are addressed, no threads remain unresolved, and the branch has no merge conflicts with `master`. - `git diff --check` and a scan of added text for secrets and private identifiers passed. ## Risks - A bridge control operation that needs more than 30 seconds now fails, even if the caller selected a longer run lifetime. Agent commands retain their own limits. - Timing out a provider promise does not cancel the remote operation or prove it stopped. Existing execution settlement still owns termination verification. No uncertain tool action is replayed. - Quiet healthy runs display warnings sooner. Existing snooze, continue, and false-positive dismissal controls still apply. Silence alone does not cancel a run, create review work, or change assignments. - No schema or API shape change. ## Model Used OpenAI GPT-6 through Codex, with tool use and code execution. The exact serving model ID and context window 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 change-specific tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 15:29:41 -07:00
Devin FoleyandPaperclip	5b8b2b38ca	feat(apps): add Neon connection (#14980) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents reach external services through the Apps catalog. Each catalog entry is a reviewed `AppDefinition` that wires a provider's hosted MCP server into Paperclip's shared vault, grants, policies, gateway, and audit trail. > - Neon is a widely used serverless Postgres provider with an official hosted MCP server, but it is not in the catalog. Teams that run their databases on Neon must use the generic "connect your own MCP server" path, which has no branding, no guidance, and no project or read-only controls. > - The connector playbook requires a catalog entry for a provider like this: the hosted server supports dynamic client registration and bearer API keys, and the common definition fields can express every option Paperclip can serialize. > - This pull request adds the Neon definition, its official artwork, the research and permission-review ledger rows, documentation, and deterministic tests, without any provider-specific runtime code. > - The benefit is a one-click, governed Neon connection with optional project pinning and read-only mode, and a documented path to live qualification. ## Linked Issues or Issue Description **Problem or motivation** Neon is a common Postgres host for the applications agents work on, but Paperclip's Apps catalog has no Neon entry. Operators who want agents to inspect schemas, run SQL, or manage branches must paste the MCP URL into the generic remote-MCP flow, which gives no branding, no provider guidance, no project boundary, and no read-only switch. **Proposed solution** Add a catalog-only Neon connection built from the connector playbook: browser sign-in through Neon's dynamic client registration with the reviewed `read` and `write` scopes, or a customer API key sent as an Authorization bearer header. Both methods expose Neon's documented `projectId` pin and `readonly` switch as optional Advanced fields. Every discovered tool stays governed by the normal per-action policies. **Alternatives considered** A plugin was not needed because no custom UI, tables, workers, or webhooks are involved. A separate read-only method was not added because the playbook treats read-only switches as advanced fields rather than methods. Neon's repeatable `category` query filter was left out because tenant fields serialize lists as one comma-joined value, so it cannot be sent correctly without new runtime code; per-action policies cover catalog narrowing instead. **Roadmap alignment** This extends the existing self-serve remote-MCP connection catalog and does not overlap planned core work. ## What Changed - Added the `neon` provider to `scripts/ingest-app-definitions.mjs` (category, API-key placement, methods, tenant fields, guidance, warnings) and regenerated `packages/shared/src/app-definitions/neon.json` plus the generated registry. - Added the Neon row to the self-serve MCP research ledger with `dcr_or_api_key` auth and risk tier S4. - Added permission reviews for `neon/mcp-oauth` (explicit scopes `read`, `write`, taken from Neon's live authorization-server metadata) and `neon/mcp-api-key` (provider key), with evidence links. - Added Neon's official tile icon (`ui/public/brands/apps/neon.png`, copied byte-for-byte from the icon linked by neon.com) and the brand manifest entry. - Added prosumer gallery copy for the Neon card. - Added `doc/connections/NEON.md` (service involvement, endpoints, administrator setup, capabilities and policy, manifest, brand provenance, validation hook) and linked it from the connections README and the permission audit. - Tests: Neon definition shape, store visibility and artwork, URL recognition, reviewed scopes with scope-widening rejection, URL projection of the project pin and read-only flag, invalid project ID rejection, the connect form's API-key gating, and the pinned catalog counts. ## Verification - `pnpm exec vitest run packages/shared/src/app-definitions.test.ts packages/shared/src/app-definitions-url.test.ts` — 34 passed. - `pnpm exec vitest run server/src/__tests__/tool-access-service.test.ts` — 367 passed. - `pnpm exec vitest run ui/src/pages/apps/AppsConnect.test.tsx ui/src/pages/apps/Browse.test.tsx ui/src/lib/app-brand-assets.test.ts ui/src/pages/apps/AppLogo.brand-assets.test.tsx` — all passed. - `node scripts/check-app-brand-assets.mjs` and `node --test scripts/app-brand-validation.test.mjs` — passed. - `pnpm --filter @paperclipai/shared typecheck`, `pnpm --filter @paperclipai/server typecheck`, `pnpm --filter @paperclipai/ui typecheck`, `pnpm check:token-gates` — clean. - Manual: in a local instance, open Apps → Browse, confirm the Neon card and icon, open `/apps/connect?source=neon`, confirm both methods, the Advanced project pin and read-only toggle, and that Connect enables after an API key is entered. The operator completed a live connection against a Neon account on this build. - Live metadata probed on 2026-10-02: both `.well-known` documents at `mcp.neon.tech` return the recorded endpoints and scopes; an unauthenticated `initialize` returns 401 with `resource_metadata`. ## Risks - Low risk to existing providers: the change is additive catalog data plus tests. The generated registry only gains one import. - Neon's hosted server grants broad project and database management. The definition carries two warnings, recommends a development project, and keeps every write under the normal action policies; the read-only switch is enforced by Neon's server, not locally. - The permission-review ledger records live proof for both methods as not run; the full lifecycle checklist in `doc/connections/NEON.md` still needs a documented pass before the entry is considered fully qualified. > 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 Fable 5.1 (`claude-fable-5-1`) in Claude Code, with extended thinking and tool use (shell, file editing, browser verification). ## 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>	2026-10-02 15:04:52 -07:00
Devin FoleyandPaperclip	d6d88b9de2	fix: preserve run outcomes when agent file cleanup is deferred (#14945) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The heartbeat service records each agent turn and releases its working files. > - A turn can save its work and finish before instruction-copy cleanup runs. > - A cleanup exception can replace that completed result with an adapter failure. > - This also loses result accounting and can prevent environment lease release. > - This pull request records a cleanup warning and keeps the original run outcome. > - The existing recovery sweep retries cleanup from the durable working-copy record. ## Linked Issues or Issue Description Related cleanup and lock work: #14866 and #14869. Related, distinct work: #14695 retains warm-process files; #12021 handles provider-process SIGTERM after a terminal result. **What happened?** An agent saved its plan, posted a comment, and requested approval. The provider completed its turn. Instruction-copy cleanup then timed out on a directory lock. Its exception escaped a `finally` block and replaced the provider result, so the completed turn showed `Run failed`. **Expected behavior** Keep the provider outcome, usage, cost, saved work, and pending approval. Record a cleanup warning and let the existing recovery sweep retry. A real provider failure must keep its original error. A failed file save must keep its failed-save receipt. **Steps to reproduce** 1. Complete a legacy adapter turn that saves work and requests approval. 2. Make instruction-copy release throw a directory-lock timeout. 3. Read the run result. Before this change, the cleanup error replaces the provider outcome. The new heartbeat tests reproduce the failure without a live provider or external service. **Paperclip version or commit** The regression reproduces on `c83df091b1a5207375eaf23466bb5c62e4e1518e`. This branch is rebased onto `cf8ad63c80`. **Deployment mode** Server-managed agent execution with persistent instruction working copies. ## What Changed - Catch instruction-copy release failures in both heartbeat teardown paths. Stop repeating a failed cleanup attempt within the same run. - Write a sanitized `instruction_cleanup` warning. A warning-write failure also preserves the run result. - Test successful, failed, and throwing providers; both teardown paths; warning-write failure; accounting; approval state; and execution-control release. - Extend the held-lock test to prove a fresh recovery worker removes the deferred copy and preserves its failed-save receipt. - Document deferred cleanup and the run-log event. ## Verification - Red proof: all five new heartbeat regression cases fail with the original release calls. - At head `20bea4f431c916d2f5db1970213aab85f5daa34c`, all 417 tests passed across heartbeat process recovery, agent directory working copies, and directory merge locks. - Full local `pnpm -r typecheck`, `pnpm build`, and `git diff --check` passed. - [GitHub CI](https://github.com/paperclipai/paperclip/actions/runs/37031119795) passed at this head. All 53 reported checks passed; the two Storybook checks were correctly skipped. This includes general and serialized tests, browser shards, runner verification, build, typecheck, and the canary dry run. - Greptile reviewed this head with 5/5, no code comments, and no unresolved review threads. The branch has no merge conflicts. - The local `pnpm test:run` attempt was stopped after it reported eight failures in unchanged suites. Four Slack/AgentMail cases selected an unrelated ancestor skills directory and failed with `ENOENT`; the two Slack cases passed with a temporary local skill-root link, which was then removed. Three company-skill cases reproduced macOS `EACCES` errors when renaming read-only cache directories. One gateway case passed when rerun alone. No full local-suite pass is claimed; the complete CI test jobs passed. ## Risks - Cleanup errors now leave recovery work pending. The durable working-copy record remains available for the existing retry sweep. - This change preserves provider failures and failed-save receipts. It does not claim that unsaved file edits were saved. - Native instruction reservation errors retain their existing behavior because they guard process ownership. - No schema, lockfile, workflow, API, or UI changes. ## Model Used OpenAI Codex, based on GPT-6, with reasoning, repository tools, and code execution. The exact backend model ID and context-window size are not exposed by 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>	2026-10-02 14:16:18 -07:00
DottaandPaperclip	9786f6df56	fix(runner): preserve credential content in document saves (#14937) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The Runner sends authorized tool calls to the control plane. > - Agents use these calls to save plans and instruction files. > - The Runner used diagnostic secret detection to reject execution arguments. > - Ordinary credential-related prose could reject a document save before persistence. > - This pull request forwards the original arguments and leaves credential policy to the provider harness. > - The benefit is reliable saves with useful diagnostic records. ## Linked Issues or Issue Description Related foundation: Refs #12415 and #14430. No duplicate save-policy fix was found. **What happened?** A `write_document` call failed before the server saved its plan. The Runner reported `semantic tool input contains credential material; refusing to execute altered arguments`. The detector also masked ordinary phrases such as `secret manager` and `credential handling` in diagnostics. Both TypeScript dispatchers had equivalent execution gates. One dispatcher also rewrote structured approval and question payloads before execution. **Expected behavior** Paperclip forwards authorized arguments unchanged. The provider harness decides credential-content policy. Log and audit redaction does not reject or rewrite save input. **Steps to reproduce** 1. Send an authorized `write_document` call with a plan that discusses credential handling. 2. Include an intentional credential value in the body to exercise harness-owned policy. 3. The old Runner rejects the call. With this change, the document service stores the exact body. 4. Diagnostic records still mask explicit credential values. Qualified credential fields, short bearer values, opaque diagnostic pairs, and valid encoded JSON token headers have regression coverage. **Paperclip version or commit** Reproduced at `c46e41e81c03cd3c8b64cf993615b604d7fe8c62`. The branch is based on current `master`. **Deployment mode** Server deployment with the native Paperclip Runner. Local regression tests use the real document service and an embedded test database. ## What Changed - Remove credential-content vetoes from Rust admission and both TypeScript semantic dispatchers. - Preserve original structured approval and question arguments during execution. - Keep transport bounds, schema checks, authorization, idempotency, and audit masking. - Require explicit credential syntax or recognized formats for diagnostic masking. Preserve ordinary prose, metadata, and dotted identifiers. - Test exact document persistence, replay, nested argument identities, and masked audit copies. - Remove obsolete retry guidance and document harness-owned credential policy. ## Verification - `cargo test --manifest-path packages/paperclip-runner/runner/Cargo.toml --locked -p paperclip-runner-core --lib --test acpx_event_payload --test acpx_provider_state --test acpx_provider_turns`: 355 tests passed. - Focused server and adapter tests: 189 tests passed after rebase. These include the real document save and the complete tool-gateway suite. - Semantic dispatcher and conformance tests: 34 tests passed. - Diagnostic redaction and MCP tests: 46 tests passed, including all six review examples. - `pnpm -r typecheck` and `pnpm build` passed on the repaired branch. - The broad local root suite was interrupted after database fixture setup failures. The focused database suites passed. CI runs the complete configured test lanes. ## Risks - Authorized tool arguments can intentionally contain credentials. The harness must enforce its content policy. - Diagnostic detection is narrower. Explicit assignments, credential fields, and recognized credential formats remain masked. - The change does not add a database migration or change company authorization. ## Model Used - OpenAI GPT-6 through Codex. The session exposes the GPT-6 model family; its exact runtime model identifier and context window size are not exposed. Used reasoning, 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 --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 14:19:41 -05:00
DottaandPaperclip	839cac1343	fix: request supported offline access for generic MCP OAuth (#14950) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents use external MCP tools through the governed gateway. > - Remote MCP connections can use OAuth access tokens that expire. > - Some providers issue refresh tokens only after an offline-access request and consent. > - Resource scopes hid that identity-provider capability in the generic connection flow. > - This pull request requests supported offline access and tests expiry through a local MCP server. > - The benefit is continued tool access without another sign-in when the provider permits refresh. ## Linked Issues or Issue Description Related PR: #13447 addresses the same OAuth symptom together with managed Codex configuration. This PR focuses on generic MCP OAuth. It also covers consent, explicit scope overrides, legacy reconnects, exact scope persistence, and real HTTP expiry tests. **What happened?** A generic MCP resource can advertise only its tool scopes. Its OAuth server can separately advertise `offline_access`. Paperclip selected the resource scopes and omitted the offline-access request. A provider could then issue an access token without a refresh token. Tool access stopped after the access token expired. **Expected behavior** Paperclip adds advertised offline access to the selected tool scopes when the OAuth server does not exclude refresh tokens. It requests consent and stores the scopes sent in the authorization request. Existing connections can discover this capability when the user reconnects. Providers without this capability keep their existing scope behavior. **Steps to reproduce** 1. Run `node scripts/mcp-fixtures/servers/oauth-refresh-fixture.mjs` from the repository root. 2. Add its MCP URL as a generic connection on a local Paperclip instance. 3. Approve the test consent page and call `read_status`. 4. Let the two-minute access token expire and call the tool again. 5. Before this fix, the connection needs another sign-in. With this fix, the call refreshes the token and succeeds. **Paperclip version or commit** The integration regression reproduced the missing-refresh-token failure on the parent of this PR's fix. The same test passes with the fix. **Deployment mode** Local development. Automated tests use a loopback HTTP MCP/OAuth server and a disposable PostgreSQL database. ## What Changed - Track offline-access capability separately from MCP tool scopes. - Add supported offline access and consent for generic connections. - Preserve the actual requested scopes through callback completion and reconnect. - Discover the capability for older connections with cached OAuth endpoints. - Add a reusable MCP/OAuth fixture with PKCE, token expiry, resource binding, and refresh-token rotation. - Test shared and personal gateway calls through two refresh rotations. Cover scope selection, unsupported refresh, and legacy reconnects. - Document the behavior and local test commands. ## Verification - The two real HTTP expiry tests failed before the fix with `oauth_refresh_missing` after the first token expired. - Tests passed on the current head: 76 generic MCP regressions, all 365 tool-access service tests, and 4 fixture controls. - Full workspace `pnpm -r typecheck` and `pnpm build` passed. Server TypeScript checks also passed after the review fixes. - All remote checks passed on `d22909bb34e9d54478c0002077c498ffe105932d`. Greptile gave 5/5 with both previous findings resolved. The complete local `pnpm test:run` is still running. - Run `node --test scripts/mcp-fixtures/servers/oauth-refresh-fixture.test.mjs` for the standalone provider controls. - Run `pnpm exec vitest run server/src/__tests__/generic-mcp-connection.test.ts` for the Paperclip integration tests. ## Risks - Users can see a consent prompt when a generic provider supports offline access. - The provider can still decline to issue a refresh token. Access works until expiry, then the user must reconnect. - A provider that advertises offline access but rejects the scope produces an OAuth error. This PR does not add an automatic retry without that scope. - Existing grants without refresh tokens need another sign-in. The fix does not change them in place. - Curated Apps keep their reviewed scope and authorization-parameter allowlists. No database migration is required. - This simulation verifies the suspected failure. The reported internal MCP server has not been tested. ## Model Used - OpenAI Codex, based on GPT-6, with reasoning, tool use, and code execution. The runtime did not expose a more specific 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 #` / `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>	2026-10-02 14:13:59 -05:00
DottaandPaperclip	7a52dcdc74	fix: repair MCP validation and cancelled execution recovery (#14951) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The tool gateway gives agents access to connected services. Recovery controls what happens when a run stops. > - Generated tool names can exceed the provider limit after the MCP client adds its prefix. > - The same invalid definition can fail each automatic retry. A cancelled run can also hold saved messages without showing its cause. > - This pull request bounds tool names, stops configuration retries, and retains cancellation evidence. > - It shows the stopped run and admits saved input only after the existing safety checks pass. > - The benefit is a clear recovery path that preserves operator Stop and prevents duplicate message delivery. ## Linked Issues or Issue Description **What happened?** A long connected MCP tool name makes the provider reject the entire request. Automatic recovery repeats the invalid request. Separately, unexpected legacy cancellations can leave saved input behind a recovery hold. The notice does not identify the stopped run or its cause. **Expected behavior** Complete MCP names fit the provider limit. Tool-definition errors require configuration repair. Cancelled runs retain their source and reason. The recovery notice shows the cause and saved-message count. Verified unexpected cancellations can start a fresh turn through the existing admission checks. **Steps to reproduce** 1. Assign an App gallery connection with a long application key and tool name to a Claude agent. 2. Start a run. The provider rejects a name over 128 characters, including its MCP prefix. 3. For cancellation recovery, stop a legacy provider turn without an operator Stop request and send a user message while the recovery hold is active. 4. Inspect the recovery notice and the deferred message queue. **Paperclip version or commit** Rebased onto master at `cf8ad63c806685bfd7c48e3ed4a919d61a7c55f1`. **Deployment mode** Hosted or self-hosted server with legacy Claude or Codex execution. Related public work: - Refs #14017. That PR caps name segments. This PR preserves existing short names and uses stable hash aliases for long complete names. It also covers classification and recovery. - Refs #4510. That PR adds a cancellation-source column. This PR records bounded evidence in the existing run result, without a migration. - Refs #12552 and #4506. Those PRs suppress recovery after operator cancellation. This PR preserves operator intent and uses the existing continuation gates. ## What Changed - Bound gateway names with the full provider prefix in the 128-character budget. Retain the original upstream tool name for dispatch and permissions. - Classify invalid tool definitions as configuration failures before diagnostic redaction. Stop automatic retries and continuation attempts for that error code. - Persist cancellation source, expectedness, initiator, reason, and time. Preserve recorded Stop intent when adapter results arrive. Report unexpected started cancellations with closed diagnostic labels. - Show the run cause, saved-message count, and Inspect run link. Offer Continue for eligible unexpected cancellations. Require verified provider stop, empty tool inventory, ownership, and the existing pause, budget, approval, and dependency gates. Use the existing queue for single delivery. - Add regression coverage and update the execution, MCP gateway, and run-log documentation. ## Verification - `pnpm -r typecheck` and `pnpm build` passed. - `pnpm check:token-gates` passed. - Ran `pnpm test:run` and completed its workspace and serialized groups. Initial resource and timing failures passed on isolated reruns. All 149 serialized route suites passed. - Reran the changed server, adapter, and UI suites after the rebase. Coverage includes long-name upstream dispatch, configuration retry suppression, cancellation evidence retention, privacy labels, oversized run projection, and concurrent saved-message delivery. - `pnpm test:e2e tests/e2e/legacy-failure-continuation.spec.ts` passed all six browser scenarios. The recovery notice shows the run cause and inspection link, and each recovery entry point reaches one new response. - Added database-backed checks for active, removed, paused, unavailable, and disabled chat connections. The final continuation and recovery-notice suites passed 167 tests. Externally bound chats hide board Continue and show a usable next action. - All 55 GitHub checks passed on `42afbf1371dcaeb72646e3d8f65c19ff7cddf8de`. Two unrelated Storybook jobs were skipped by their normal conditions. Greptile reviewed that commit at 5/5 with no findings and no open review threads. ## Risks - Long tool names change to aliases. Existing short names stay compatible. The original connection and upstream name remain the dispatch authority. - Invalid tool definitions no longer get automatic retries. An operator must repair the configuration before a new attempt. - Continuation changes apply only to positively identified unexpected legacy cancellations with complete empty tool inventory. Operator Stop, unknown historical cancellations, outstanding tools, and unverified provider termination keep their holds. - No database migration. The added projection fields are optional. Cancellation reason and initiator IDs remain local run evidence; Sentry receives only closed source and initiator-type labels and expectedness. ## Model Used - OpenAI GPT-6 through Codex, with reasoning, repository editing, shell execution, and GitHub tool use. The runtime does not expose the exact model 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>	2026-10-02 13:47:59 -05:00
DottaandPaperclip	7d59de6113	feat(connections): probe provider usage limits on demand (#14936) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Connections store the AI accounts used by legacy and native runners. > - Subscription accounts can reach session, weekly, model, or paid usage limits. > - Operators need to read these limits for a specific stored account before making a routing decision. > - This pull request adds an on-demand usage probe to the connection service and account detail. > - The result preserves provider limits, reset times, paid usage, and unknown values for later consumers. ## Linked Issues or Issue Description **Subsystem affected** Shared contracts, the connection service and API, and the account detail UI. **Problem or motivation** Managed AI accounts lack a common operation to read their current usage limits. A local harness probe can read a different login from the account selected for an agent. **Proposed solution** Add `aiConnectionService.probeUsage()` and a board-only connection usage endpoint. Probe the selected credential grant on request. Support Codex, Claude, and Grok subscriptions, plus OpenRouter API key limits. **Alternatives considered** Harness-specific automatic polling would couple the read to execution and can read ambient credentials. This change uses the managed connection credential and leaves scheduling and admission decisions to later work. **Roadmap alignment** This extends the existing Personal & Shared AI Accounts capability. It adds no routing or quota enforcement. Related: Refs #14459 for managed OpenAI quota reads; Refs #14781 and Refs #13379 for downstream pacing and budget work. This operation reads one requested account across all three subscription providers. ## What Changed - Add typed usage snapshots and a probe capability flag to managed AI connections. - Normalize Codex, Claude, Grok, and OpenRouter responses. Keep model scopes, provider admission, reset periods, and paid allowances separate. Preserve unknown values. - Enforce company membership, credential audience, grant identity, and connection lifecycle before reading the stored secret. - Add a board-only `GET /api/companies/:companyId/ai-connections/:connectionId/usage` endpoint with `no-store` responses. - Add manual **Check usage** and **Refresh** actions to account details. Show compact usage bars, resets, admission and overage status; remove repeated descriptions and account-default copy. Clear previous results during a new request or error. - Add Storybook previews using the production account components for all four providers, initial checks, loading, and permission errors. - Add provider, authorization, runner selection, API, and UI coverage. Document provider sources and live qualification. ## Verification - Initial provider, authorization, selection, API, and UI validation passed (96 focused tests): `pnpm exec vitest run server/src/services/ai-connection-usage.test.ts server/src/__tests__/ai-connections.test.ts ui/src/components/ai-connections/AiConnectionUsagePanel.test.tsx server/src/__tests__/openapi-routes.test.ts`. - `pnpm -r typecheck` passes for the initial implementation. After simplifying the UI, 9 usage-panel and date-helper tests, UI typecheck, token gates, and Storybook build pass. The initial feature module boundary check also passed. - Real Codex, Claude, and Grok credentials were saved to encrypted disposable connections. The actual usage HTTP route returned 200 with `status: ok`. Legacy and native runner selection checks passed. The tests started no model turn and exchanged no refresh token. The disposable databases and vaults were removed. - Live Claude responses added structured scoped limits. Live Grok responses omitted included-plan usage. Tests now cover both shapes and preserve the Grok omission as unknown. - The full workspace build passes. A full local test run hit a heartbeat feedback timeout. That case passes in isolation. The duplicate local run was stopped after all remote checks passed. The Slack ordering and OpenCode transport CI flakes also pass in isolation and on the CI rerun. - Current head: `ff3d479029a1c4248190323e221b2803cfb0d79d`. All 54 active checks pass. Two Storybook checks are intentionally skipped by the workflow. Greptile is 5/5 with no unresolved review findings; the branch is mergeable. ## Risks - Subscription usage endpoints can change. Credentials can lack usage-read permission. The probe returns explicit errors without fresh limits in these cases. - A successful probe can contain partial data. Missing utilization or admission remains unknown. An enabled paid-usage switch does not prove a funded balance. - This change adds no migration. It does not change runner admission or automatic provider selection. Provider requests use fixed endpoints, disabled redirects, bounded response sizes, and a 15-second deadline. ## Model Used OpenAI Codex, GPT-6, with reasoning, file editing, shell execution, and HTTP tools. The session does not expose the exact runtime model variant or context window size. Real provider credentials were used only for the authorized live checks. ## 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>	2026-10-02 11:47:14 -05:00
DottaandPaperclip	6c1a75da49	feat(connections): make AgentMail a default connection with inline setup (#14772) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Connections give agents access to external services. > - AgentMail needs both a saved key and an inbox assigned to the agent. > - Chat requests offered a setup link instead of an inline card and could treat a saved key as complete. > - Inbox setup also hid address conflicts behind a generic server error and a separate review step. > - This pull request makes AgentMail a default connection, adds the inline card, reduces setup to two steps, and shows conflicts beside the address. > - Shared native dropdown styles also give every caret a consistent inset. ## Linked Issues or Issue Description **What happened?** AgentMail requests in chat did not show a usable inline connection card. Manual setup required extra screens, ignored saved account keys, and could trap new-address setup in a locked inbox dropdown. Agent selectors omitted the avatar from the selected value. A taken address could produce an HTTP 403 from AgentMail and appear as an internal server error. Native dropdown arrows also touched the right edge of their fields. **Expected behavior** Make AgentMail available as a default connection. Ask for the API key inline, with a direct link to its provider page. Default human access to the company and agent access to the requesting agent. Resume the agent only after an assigned inbox is active. Manual setup should ask for an agent and email address, then finish. Address checks should run as the user types. Taken addresses should show clickable alternatives. A domain dropdown beside the name should prefer a verified custom domain. Setup should suggest authorized saved AgentMail keys and show agent avatars in the picker and selected value. **Steps to reproduce** 1. Ask an agent to connect AgentMail when it has no assigned inbox. 2. Check that an inline API-key card appears and links to the provider's API-key page. 3. Open AgentMail setup, choose an agent, and request an address that is already taken. 4. Correct the inline error, refresh, and finish setup with the same request ID. 5. Inspect native dropdown carets in light, dark, disabled, and right-to-left states. Uses the bounded provider-error parser merged in #14768. Related work: #13256 introduced AgentMail; #14725 expanded connection search. ## What Changed - Stop recurring email queries for tasks that have no email thread. Share the query between the thread provider and activity view. Keep email-task updates and invalidation-based discovery. - Make AgentMail available without the experimental chat setting. Keep the catalog, setup and management routes, agent Channels tab, task email feed, receiving worker, and agent tools available by default. Other experimental chat providers stay gated. - Make the email address and copy icon a single clickable action with the shared Copied! confirmation. Add View inbox linking directly to the matching AgentMail console inbox, with the address encoded as one URL path segment. - Reorganize inbox Settings around the copyable email address, usage instructions, and receiving status. Move reconnect credentials into a disclosure and separate the Disconnect action. Add production Settings stories for active, paused, unassigned-address, revoked, webhook, long-address, mobile, and reconnect states. Show repair controls when the inbox has an error. Keep usage instructions tied to an active inbox with an address. - Add AgentMail channel intents and an inline key field with the direct API-key URL. - Keep setup and retry state tied to the interaction. Require an active inbox for completion. Preserve company and agent access checks. - Reduce manual setup to agent selection and email selection. Put the domain dropdown beside the address and default to a verified custom domain. Preserve explicit choices across reloads. Keep receiving settings under Advanced options. - Check the initial address and edits after a 350 ms pause. Abort superseded requests and ignore stale responses. Show clickable suggestions and retain known creation conflicts across reloads. - Add a company-scoped, manager-only address check using the saved credential. Search the visible inbox list instead of fetching an uncreated inbox: live AgentMail retains negative lookups that can break subsequent access-key creation. Unlisted addresses remain unknown; creation is authoritative. - Suggest labeled saved AgentMail keys in both manual setup and the inline card. Filter by company, provider, active credential, and current-user grants on the server. Prefer an account key and preserve the selected key or an explicit new-key choice across refresh. Use verified scope metadata and bounded concurrent checks for legacy keys. Never return secret values. - Catch an inbox-only key before the email step. Allow its existing inbox only after an explicit choice. Recover old locked drafts at the key picker. Save the replacement key before retiring an empty draft, then use a new setup URL so refresh preserves the switched account; stop if cleanup fails. Preserve already allocated addresses and their original accounts. - Use the shared AgentSelect in email setup. Show the canonical agent avatar in each option and the selected value, including other consumers of the shared component. Add regression coverage for legacy and current Lucide agent-mention icon formats. - Start each catalog Add connection with a fresh setup identity. Honor Finish setup's exact draft/account/address instead of resuming an unrelated browser draft. Return Cancel and Done to Connectors and Email settings to the inbox. Group the task/thread explanation in a How it Works card. - Route AgentMail catalog removal through the email inbox control API, including unfinished drafts. Refresh both the catalog and inbox views. - Render each inbox management tab separately. Access uses the saved account grants and agent controls; Conversations and Activity use the shared persisted email feed. Activity lifecycle actions use the email API. Reconnect returns to inbox Settings. Conversation failures show a retry instead of a false empty state. Email delivery recovery stays in the task. - Map documented provider address conflicts to a field error. Preserve actionable messages for other failures. - Preserve non-secret draft fields across refresh, scoped to the requested agent. Never save API keys in browser storage. Resume partial inbox creation with the original agent, address, and request ID. - Show an already-created address with explicit retry and new-address recovery instead of locked inputs. Preserve the original inbox and resumable draft when choosing another address. Distinguish runtime-key 404 errors and log safe provider status/operation/code. - Apply final agent access once within email setup authorization for a new account whose original installs are unchanged. Preserve later permission edits and reused account installs. Support in-place retry of progress loading. - Let a failed inline setup change keys after retiring an empty draft. Persist its replacement setup identity without storing secrets. Recover a server-saved account when refresh interrupts the save response, while preserving intentional account changes. - Render the production setup in Storybook and add error, recovery, and mobile states. - Inset native select carets in shared CSS. Preserve custom icons, listboxes, keyboard behavior, and forced-color controls. - Add browser regression coverage and an AgentMail Product E2E case with persisted-state and rendered-card evidence. ## Verification - Full `pnpm -r typecheck`, `pnpm build`, `pnpm check:token-gates`, and `git diff --check` passed after the default-availability change. - All 485 focused tests passed. These cover setup, management, catalog and route gates, connection intents, email authorization, Cursor execution, and the OpenAPI contract. All 39 email integration tests run with the experimental chat setting off. - The shared polling change passed four behavioral tests, UI typecheck and build, and token gates. - `tests/e2e/agentmail.spec.ts` passed with the actual server setting off. This full-stack browser test uses simulated provider responses. It covers catalog entry, saved keys, editable address and domain controls, creation, conflicts, retry, all management tabs, clipboard feedback, the provider link, and task email rendering. - In the live local browser, Add connection reached the editable email step with the saved account key. The verified custom domain was selected by default. Both domain choices worked. The existing inbox Settings page remained available. Both active inboxes completed new mail checks with the setting off. No new provider inbox or email message was created for this pass. - Earlier live provider acceptance covered creation on a verified custom domain, Finish connecting on the reported draft, successful mail checks after refresh, and catalog removal of disposable draft and active connections. Clicking the email address copied the exact address and showed Copied!. View inbox opened the same inbox in AgentMail’s console. No email messages were sent. - Production setup and Settings Storybook builds and interactions passed. Settings states include active, paused, unassigned, revoked, webhook, long-address, mobile, and reconnect. Receiving and revoked-access stories had zero accessibility violations. - Full local `pnpm test:run` on an earlier revision completed with 14,709 passing, 87 skipped, and four transient failures. All four failed cases passed in focused reruns without product changes. That serial full local command was not repeated after each follow-up. The latest-head full CI suite is the final test gate. - CI found an obsolete browser assertion that hid every channel when the flag was off. Updated it to keep AgentMail and the Channels surface visible while preserving the GitHub chat route gates. All 11 provider browser tests passed locally after scoping the Channels selector to the agent sidebar. Two initial local attempts stopped at temporary Postgres initialization. The passing run used a separate disposable database on the existing local Postgres server; it was removed after the test. - Updated the remaining sidebar and aggregator discovery assertions for default AgentMail availability. Ordinary task fixtures now return no email thread. All 128 sidebar/task-page tests and all 42 aggregator tests passed locally. - Latest head `b42bb4cd5cc9f2d01a99ab8026832d5a956ea85f`: full CI passed, with 54 successful checks including Snyk and two intentional Storybook skips. The CI run is https://github.com/paperclipai/paperclip/actions/runs/37020833647. A fresh Greptile review scored 5/5 with no unresolved threads. Live model evaluations and inbound/outbound email delivery were not run. ## Risks - AgentMail no longer needs experimental opt-in. Setup still requires a human to connect an account and assign an inbox. Inline setup creates an inbox after a human submits a new or saved key. Company access, agent access, inbox assignment, and completion checks remain enforced. - AgentMail read APIs cannot prove global address availability. The visible-list check is bounded to 100 entries and cannot see inboxes outside the key’s scope. The UI reports this limitation, suggests alternatives without claiming they are free, and keeps final creation conflicts inline. Lookup outages show an error without preventing the authoritative creation attempt. - Native select CSS affects the whole app. Custom-icon selects and multi-row lists are excluded. Forced-color mode keeps the browser caret. - Saved-key discovery uses stored verified scope metadata and checks authorized legacy credentials concurrently within a shared three-second deadline. Provider outages mark legacy choices unavailable; users can still enter another key. Final use rechecks authorization and provider access. - No database migration or transport default change. Live connection remains the default. ## Model Used OpenAI Codex, GPT-6, with reasoning, tool use, and code execution. 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; full-suite limitation 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 (latest head `b42bb4cd5cc9f2d01a99ab8026832d5a956ea85f`) - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups (latest head `b42bb4cd5cc9f2d01a99ab8026832d5a956ea85f`) - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 10:01:15 -05:00
DottaandPaperclip	ec3bacc9bd	fix(chat): hide ignored provider information (#14929) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Task and agent chats show agent progress and problems that need attention. > - Codex also sends account, skill, and unrelated thread notifications. > - The runner correctly ignores that information but reports it as a warning. > - Chat then shows an internal diagnostic as an actionable provider notice. > - This pull request keeps the diagnostic in run logs and removes it from chat. > - Real provider warnings, errors, and agent replies remain visible. ## Linked Issues or Issue Description **What happened?** Chat showed “Received a provider update” and a warning with the text “ignored unrelated provider information”. Its details said “User Actionable: Yes” even though no user action was needed. Saved conversations retained the same noise. **Expected behavior** Keep ignored provider information in the run log. Do not show it as chat activity or a user warning. Preserve real warnings and errors. **Steps to reproduce** 1. Start a conversation with the native Codex runner. 2. Have the provider send an account update, skill change, or unrelated thread notification during the turn. 3. Inspect live chat and reload its saved history. The regression tests also reproduce the old stored notice without a live account. **Paperclip version or commit** Source implementation on master at `e00d10d5d`. The duplicate search found no open PR for this fix. Related prior work: #13109 improved provider-notice presentation. #12367 added Codex thread normalization. This change addresses the internal information that those paths still projected as chat warnings. **Deployment mode** Native Paperclip Runner with the Codex app-server provider. The issue was seen in hosted chat and can be reproduced with local provider fixtures. ## What Changed - Map ignored unrelated Codex information to `harness.diagnostic` in the Rust and TypeScript normalizers. - Retain a bounded allowlist of redacted provider method and thread/turn identifiers. - Use the same Unicode character limit and truncation marker in both normalizers. - Share the text redactor through a pure helper. Keep provider connection code out of the standalone demo's source closure. - Omit that diagnostic and the matching legacy notice from live chat. - Omit the matching legacy notice from saved chat history. - Test diagnostic retention, account-notification integration, live and saved chat, and continued visibility of real warnings, errors, and replies. - Document the local run-log event and historical display behavior. ## Verification - Passed: 68 tests in the two affected UI transcript suites. - Passed: 60 TypeScript tests across provider events, transport behavior, and the standalone demo boundary. - Passed: 13 Rust provider-event tests and the Codex account-notification integration test. - Passed: `pnpm check:token-gates` and Cargo formatting checks. - Passed: full `pnpm build` and `pnpm -r typecheck`. After the review fix, the provider package build, typecheck, and both provider-event suites passed again. - Full local `pnpm test:run` failed: 608 files / 10,904 tests passed, 30 server suites failed, and 104 files / 4,012 tests were skipped. Most failures were embedded PostgreSQL startup errors. Two tests timed out in `heartbeat-comment-wake-batching` and `workspace-git-snapshot-streaming`. PostgreSQL startup also failed in `heartbeat-run-event-sequencing` and `native-finalization-migration`. These server files are unchanged by this PR. Isolated heartbeat reruns were skipped locally. The stable test script stopped after this general-server group, so later groups did not run locally. - The original review thread is resolved. Greptile is 5/5 on current head `683dab7cce57187c57e84c83f5e9da4ad75c9c04`. - All current-head CI gates passed, including the full server/chat/workspace test matrix, Rust and TypeScript runner suites, browser E2E, build, typecheck, and release canary. [CI run](https://github.com/paperclipai/paperclip/actions/runs/37021330663). - Replay the exact old warning in either transcript adapter. It must produce no chat row. A genuine provider warning or error must still produce a row. ## Risks - Low risk. The display filter matches one diagnostic code or the complete legacy warning shape. Other provider notices remain visible. - New ignored-information events use the existing harness-diagnostic event type. They retain diagnostic evidence without original account payloads. - No database migration, API permission, provider execution, or recovery behavior changes. This affects the local run log, not Telemetry or OpenTelemetry exports. ## Model Used OpenAI Codex, GPT-6. The exact backend model ID and context-window size are not exposed in this session. Used reasoning, repository inspection, code editing, tool use, and 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 #` / `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 affected tests locally and they pass (the broad local run has PostgreSQL startup errors and timeouts documented above; the full CI matrix passed) - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 09:59:34 -05:00
DottaandPaperclip	e00d10d5d5	fix(connections): repair stale AI defaults from agent settings (#14916) ... ## Thinking Path > - Paperclip manages AI agents and controls the credentials used for their work. > - Managed AI connections resolve each responsible user's provider default. > - Agent settings created another account but kept the old default selected. > - A rejected provider test left the old account marked as connected. > - Claude ACP reported a typed login failure as a generic terminal-access error. > - This pull request repairs the selected account or selects the new login explicitly. > - Agents can save and run with the repaired credential, and failed logins request sign-in. ## Linked Issues or Issue Description - Fixes #14831. - Refs #13867. Environment failures remain separate from credential-health failures. ## What Changed - Add an agent-settings action to reconnect an unavailable personal default in place. Keep its connection, grant, default, and agent access. - State that a new account becomes the user's provider default. Select its returned grant before changing the agent binding. Keep the actual sign-in method. - Show default-update errors and allow retry without another provider login. - Show the agent-access choice. Connection managers start with company-wide access for their own tasks. Other members start with access for the current agent. - Use the server's connection-manager permission in the shared list response. This includes members with a custom management grant. - Mark credentials as needing attention after an explicit login rejection in Test or Save. This includes API-key 401 and 403 responses. Network, quota, and server failures keep the credential health unchanged. - Reuse the credential-generation check so an old failure cannot invalidate a newer reconnect. - Route Claude's typed provider `access` failure to the existing login-recovery flow. Replace its generic terminal-access fallback with a sign-in message. - Add regression tests and update the AI Connections documentation. ## Verification - Red: the UI tests failed on the missing reconnect action, unused returned grant, missing access choice, and lost default-update error. The server tests failed because rejected credentials stayed connected. The real ACP fixture returned `acpx_turn_failed` for typed login failures. - Green: 156 tests passed across the AI connection, hiring, agent field, and New Agent suites. All 37 environment-route tests passed. The Claude ACP authentication fixtures also passed. - `pnpm check:token-gates` passed. - `pnpm -r typecheck` passed. - `pnpm build` passed. - The full local `pnpm test:run` passed 707 files and 14,503 tests, then exited with an agent-conversation timeout and embedded PostgreSQL startup failures in unchanged suites. The isolated conversation and migration tests passed on rerun. Later local test groups did not run after this failure. - [All CI gates passed](https://github.com/paperclipai/paperclip/actions/runs/37012669356) on commit `38513dfe2`. This includes the full test matrix, browser tests, typecheck, build, Runner checks, and canary dry run. - Greptile reviewed commit `38513dfe2` and returned 5/5 with no open findings. - The regression tests use a real embedded database and a real ACP fixture process. Live provider sign-in requires a valid account and was not run. ## Risks - Connecting a new account from agent settings changes the user's provider default. The dialog states this before sign-in. - The displayed access choice can allow all company agents to use the account for its owner's tasks. Reconnect keeps the existing access. Server permissions still control installs. - Claude's typed `access` category maps to the provider's `auth_required` signal. Tool and workspace request failures retain their existing classification. - No database migration or provider credential format changes are required. ## Model Used - OpenAI GPT-6 through Codex. The exact served model identifier and context window are not exposed in this session. Capabilities used: reasoning, repository tools, code editing, and command 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 --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 08:57:52 -05:00
DottaandPaperclip	408f70e69f	fix(runner): preserve stock Codex base instructions (#14920) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The native Runner connects Paperclip tasks to Codex app-server. > - Paperclip passed its runtime context as `baseInstructions`. > - That field replaces the stock Codex base prompt. > - This pull request sends the same Paperclip context as additive developer instructions. > - Codex keeps its stock prompt and still receives Paperclip task instructions and tools. ## Linked Issues or Issue Description **What happened?** The native Codex driver and Rust provider sent Paperclip context through `baseInstructions` on thread start and resume. Codex used this text in place of its stock base instructions. Direct-chat resume also sent an empty replacement base. The Runner Lab session path used the same replacement field. **Expected behavior** Codex should retain its stock base prompt. Paperclip should add its runtime context through `developerInstructions`. Other provider facades should retain their current instruction handling. **Steps to reproduce** 1. Create a native Codex session through Paperclip Runner. 2. Inspect the `thread/start` request in the native provider trace. 3. Resume the session and inspect `thread/resume`. 4. Before this fix, these paths set `baseInstructions`. After this fix, the Codex paths set `developerInstructions` and omit `baseInstructions`. **Paperclip version or commit** Reproduced against master at `cad26c6bfb736039c8ed5743da650a44792a083c`. **Deployment mode** Built from source. Native Codex app-server and runnerd paths. A local protocol probe used codex-cli 0.153.4 and a localhost Responses stub. No duplicate fix or matching public issue was found in the GitHub search. ## What Changed - Send additive developer instructions on Codex start and resume in the TypeScript driver, Rust provider, and Runner Lab session path. - Carry the additive fragment through runnerd, including runtime asset path mapping. - Preserve existing instruction fields for other provider facades, including OpenCode. - Add start/resume/direct-chat regression coverage and check the actual Rust provider request. - Document the historical option and trace field names. Record progress and follow-ups in the working checklist. ## Verification - `pnpm -r typecheck` — passed. - `pnpm build` — passed. - Targeted Codex driver lifecycle, driver, and live-session Vitest suites — 139 tests passed. - `cargo test --manifest-path packages/paperclip-runner/runner/Cargo.toml --locked -p paperclip-runner-core --test codex_provider` — 91 passed, 2 ignored subprocess helpers. - Real app-server probe: a localhost Responses stub captured identical 14,732-character stock base instructions on fresh start and cold resume. Both requests retained the Paperclip marker in developer input. Both stub turns completed. No paid inference was used. - Runnerd transport Vitest suite — 182 tests passed. - The initial `pnpm test:run` attempt reported local dependency-loading, embedded PostgreSQL startup, and macOS `/var` versus `/private/var` path failures. It was stopped after those failures. Loading-suite reruns passed 1,428 tests after the build; native interaction/finalization reruns passed 38 tests. A seven-suite diagnostic rerun passed 463 tests and isolated the remaining path and PostgreSQL setup failures. - With `TMPDIR=/private/tmp`, workspace, gateway, interaction, and attachment suites passed all 356 tests. The remaining environment-image and native-session-resumption suites passed all 44 tests with the same canonical temp path. All affected suites passed on rerun. The original full local command was stopped after failures and is not claimed as passing. - All 55 PR checks passed at `83281439456181396f3707eecda5d2ebc90bd14d`. Greptile scored 5/5 with no open review threads. - No paid live campaign or Product E2E browser suite was run. This change has protocol and regression coverage; it does not claim improved task quality. ## Risks - Stock Codex behavior may differ from behavior under the previous Paperclip replacement prompt. Restoring that behavior is the intended change. - Existing Codex threads retain their saved replacement base prompt. They need a provider session reset to receive the stock base. This PR does not reset active sessions or alter recovery rules. - The legacy `baseInstructions` option and trace field names remain for compatibility. They now describe the additive Paperclip fragment for Codex. - The separate Codex-through-ACP dependency patch remains a follow-up in the harness coverage checklist. This PR covers native app-server execution. ## Model Used OpenAI Codex, GPT-6. The exact runtime model variant and context window are not exposed in this session. Used reasoning, repository inspection, code editing, shell execution, and test 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>	2026-10-02 08:53:47 -05:00
DottaandPaperclip	c46e41e81c	fix(heartbeat): validate native MCP gateway ownership (#14914) ... ## Thinking Path > - Paperclip lets people manage agents and govern their tool access. > - Native runs receive an immutable MCP tool assignment for one agent. > - The gateway must enforce that owner when it authenticates a run token. > - Older gateway rows stored the owner only in metadata. > - This change validates the gateway and profile, binds new rows, and repairs valid older rows on reuse. > - It also delivers each native assignment once. > - Other agents cannot use the assignment, and explicit shared gateways keep their configured scope. ## Linked Issues or Issue Description Builds on #14012 by @busla (Jón Levy). That PR adds agent binding and seven regressions. This PR carries that fix onto current master and adds legacy authentication, profile validation, and duplicate-delivery coverage. Related: #14864 improves discovery memory use. **What happened?** Native gateway creation stored an owner in metadata but left `agentId` null. Authentication could therefore accept another agent's run token. Managed discovery could also deliver historical native assignments again. **Expected behavior** A native assignment accepts only its owner's run token. The gateway and profile must refer to the same immutable assignment. The current assignment enters the run configuration once. **Steps to reproduce** 1. Run the database fixtures in `heartbeat-runtime-mcp-servers.test.ts` on the baseline. 2. Create a native assignment and inspect its stored gateway owner. 3. Authenticate with another agent's run token, then inspect legacy reuse and managed delivery. 4. The baseline fails six ownership and delivery cases. The fix passes all twelve cases. **Paperclip version or commit** The red baseline is `f2e0f1963`. This PR is based on `cad26c6bf`, which includes the merged discovery fix. **Deployment mode** Native Paperclip Runner execution and managed Codex MCP delivery. Reproduction uses isolated database and HTTP fixtures. ## What Changed - Store the agent owner and agent context on new native gateways. - Validate profile and gateway assignment metadata before reuse or token creation. - Bind valid legacy rows with a company-scoped, null-owner update and validate the result. - Reject mismatched run tokens before legacy repair. - Identify native assignments by gateway metadata, the reserved profile key, or profile source. Reject missing or malformed provenance, including JSON null. - Exclude historical native assignments from managed gateway delivery. Keep their rows for existing runs. - Add twelve database and HTTP regressions and document the runtime contract. ## Verification - Red baseline: six regressions fail and four controls pass before the initial fix. Two additional regressions reproduce metadata-loss admission and a JSON-null TypeError before the review fix. - All twelve ownership regressions pass on the final code, including owner admission, cross-agent rejection, metadata loss, JSON-null HTTP 401, and explicit shared-gateway admission. Policy, listing-memory, and discovery HTTP coverage also passes. - Full workspace typecheck and build pass locally. Server typecheck and compilation pass again after the review fix. The final ownership and grant patches pass 42 combined database and HTTP regressions. - [Full CI](https://github.com/paperclipai/paperclip/actions/runs/37011383657) passes for `626a08ae66361cf586105877e24d806b1a7a9c20`: all 54 checks succeed; two optional Storybook checks skip. This includes full typecheck, build, all test shards, all eight E2E shards, runner verification, and the canary dry run. - Greptile scores that exact head 5/5. No review threads remain unresolved. ## Risks Invalid historical native gateway or profile metadata now rejects authentication. Valid unbound rows are repaired only when their owner reuses the assignment. Conflicting owners are never overwritten. Historical rows are retained for existing runs. Explicit shared gateways use ordinary profiles and keep their configured scopes. The reserved native profile namespace remains agent-owned even when gateway metadata is cleared. No schema or dependency changes are included. ## Model Used Original fix and seven regressions in #14012: Anthropic Claude Opus 5.5, `claude-opus-5-5`, 1M context, as reported by @busla. Extensions and verification: OpenAI Codex (GPT-6), with reasoning, repository inspection, code execution, and tests. This session does not expose the exact serving model identifier or context window. ## 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>	2026-10-02 08:22:33 -05:00
DottaandPaperclip	483dbc8890	fix(tool-access): enforce stored grant restrictions (#14915) ... ## Thinking Path > - Paperclip governs the tools that agents can discover and call. > - Stored grants can limit access to a tool, connection, or application. > - The grant matcher must enforce every restriction in that scope. > - A nonmatching allow list fell through to a policy selector matcher that ignores allow. > - This change requires an explicit allow match and validates additional selectors. > - Malformed and unknown restrictions deny access. > - Discovery and execution now enforce the same stored grant limits. ## Linked Issues or Issue Description Related: #14864 adds the shared database and HTTP discovery fixture used here. Searched existing public PRs for tool grant scope fixes. No duplicate scope-validation fix was found. **What happened?** A stored grant with a nonmatching `scope.allow` could authorize a tool. Empty or malformed allow lists, unknown selectors, and combined mismatching selectors could also authorize access. The fallback policy matcher does not validate stored grant JSON. **Expected behavior** An explicit allow list must match the requested tool, connection, or application. Every additional selector must also match. Unknown or malformed restrictions must deny access. Existing null and empty-object scopes keep their broad grant behavior. **Steps to reproduce** 1. Run `tool-grant-scope.test.ts` on the baseline. 2. Create a deny profile and a grant that names another tool. 3. Attempt discovery or a call for the tool outside the grant. 4. The baseline authorizes access. The fix denies it. **Paperclip version or commit** The red baseline is `f2e0f1963`. This PR is based on `cad26c6bf`, which includes the merged discovery fix. **Deployment mode** The company-scoped MCP gateway. Reproduction uses isolated database fixtures and a deterministic HTTP provider. ## What Changed - Require an explicit allow entry to match the gateway or upstream tool name, connection, or application. - Apply all additional selectors after the allow match. - Reject unknown selectors, invalid value types, empty restrictions, and non-object scopes. - Preserve null and empty-object scope compatibility. - Add sixteen regressions, including discovery, successful execution, and revocation through the HTTP gateway. - Document stored grant scope behavior. ## Verification - Red baseline: seven restricted-scope cases and three malformed-root cases fail. HTTP discovery also exposes tools outside the grant. - All 16 grant regressions and 35 adjacent policy tests pass locally. The HTTP test excludes an ungranted tool from discovery, returns 403 for its call, and verifies that no provider call occurs. It also checks successful execution and later revocation. - Server typecheck passes. The final ownership and grant patches also pass 42 combined database and HTTP regressions. - [Full CI](https://github.com/paperclipai/paperclip/actions/runs/37011177989) passes for `803fa9440111742672c94c4471e5b98f15dd3b97`: all 54 checks succeed; two optional Storybook checks skip. This includes full typecheck, build, all test shards, all eight E2E shards, runner verification, and the canary dry run. - Greptile scores that exact head 5/5. No review threads remain unresolved. ## Risks Stored scopes with unknown keys or malformed restrictions now deny access. Operators must correct those grants before they can authorize tools. Null and empty-object scopes keep their previous broad behavior. There are no schema, dependency, or API changes. ## Model Used OpenAI Codex (GPT-6), with reasoning, repository inspection, code execution, database regressions, and HTTP tests. This session does not expose the exact serving model identifier or context window. ## 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>	2026-10-02 08:21:40 -05:00
4 people	cad26c6bfb	fix(tool-gateway): bound MCP discovery memory and concurrency (#14864) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents discover governed tools through the MCP gateway. > - A listing repeated policy and full run-row reads for each catalog tool. > - Parallel listings multiplied those allocations during run startup. > - One 900-tool baseline listing used 11,489 queries and about 1.7 GiB of extra heap in a fixture. > - This pull request shares reads within a listing and bounds whole listings across the process. > - The benefit is lower discovery memory use while execution still checks current policy. ## Linked Issues or Issue Description Refs #13115. Its on-demand target change affects the same listing loop. **What happened?** MCP discovery repeated roughly 13 reads per tool. Full run snapshots and repeated connection configurations caused large allocations. Per-listing bounds alone did not limit concurrent listings across gateways. **Expected behavior** Discovery reads shared inputs once per listing. The process bounds active and queued listings. Catalog payload size and policy evaluation still grow with the catalog. Tool execution checks current access rules. **Steps to reproduce** Create a company with a remote MCP connection, 900 catalog tools, large schemas, and a large run snapshot. Send concurrent tools/list requests using a run-bound gateway token. Run the committed benchmark for a deterministic reproduction. **Paperclip version or commit** Baseline: f2e0f19630. Measured on Node 26.4.0 on macOS. ## What Changed - Keep Michael Nguyen's per-listing policy cache, scalar policy reads, 16-decision bound, and catalog/connection split under repeatable read. - Admit two whole listings per process and queue at most 32. Return 503 with tool_discovery_busy when full. - Propagate disconnects to discovery. Stop scheduling new reads and drain started reads before freeing the slot. - Project context IDs in gateway authentication and GitHub runtime discovery. Avoid loading task descriptions and results. - Keep discovery audit counts and a SHA-256 digest instead of the full name list. Retain access and activity audit records. - Sweep expired named gateway tokens at startup and on the existing scheduler. Delete at most 500 per pass. Add an idempotent expiry-index migration. Resolve audit token references atomically so cleanup cannot break admitted requests. - Return GET 405 with Allow: POST on stateless MCP gateway and runtime-tools endpoints. Keep runtime-tools authentication. - Add red-green regressions, HTTP concurrency and policy-revocation coverage, and browser approval assertions. - Commit the benchmark harness, raw results, and resource-bound documentation. ## Verification - Baseline listing regressions failed with 722 queries for 50 tools and 6,422 for 500. The connection-row duplication regression also failed. - Follow-up regressions failed before the fixes for full snapshot reads, abandoned listings, full name-list audits, and expired tokens. - Listing and scheduler tests: 14 pass. Existing gateway/policy suites passed after preserving the cleanup return contract. - Two HTTP journeys pass: initialize, GET/SSE rejection, 16 concurrent 500-tool listings, provider call, policy revocation, and denied retry. Token cleanup during provider dispatch also completes successfully and blocks subsequent requests. - pnpm test:e2e tests/e2e/mcp-user-stories.spec.ts --grep '@mcp-runnable': 8 pass. The approval journey clicks Allow once in the browser. Review screenshots wait for loaded content. - pnpm -r typecheck: passes. pnpm build: passes. - The general server group completed with 14,755 passes and 18 failures. The 17 startup mock failures were fixed; all 21 startup tests pass on rerun. The one Discord timing failure passed on the unchanged baseline and on rerun (74 tests). UI and CLI groups pass 7,632 tests. Shared and skill groups pass 853 tests. The remaining database and adapter groups pass 3,010 tests with one worker after a macOS shared-memory limit interrupted a parallel run. All 149 serialized route files pass (2,762 tests). - At 900 tools, one listing falls from 11,489 to 36 queries and from about 1.7 GiB to 37 MiB of extra heap. Sixteen concurrent listings used 169–187 MiB of extra heap. Four connections used 39 queries per listing. - Latest-head verification: 54 successful checks and two expected skips on 5c0793090c. Fresh Greptile review: 5/5 with no unresolved threads. - Reproduce with server/scripts/benchmark-tool-gateway-listing.ts. See doc/mcp-discovery-performance.md and doc/benchmarks/2026-10-01-mcp-discovery.json. ## Risks - The process-wide FIFO queue can increase discovery latency. Excess callers must retry 503 responses. - Cancellation applies to discovery. Started database reads finish before their slot is released. - Audit consumers must use visibleToolCount and visibleToolsHash instead of visibleTools. - The expiry index can briefly lock the token table during migration. Sweeps preserve unexpired and non-expiring tokens. - Measurements use isolated fixtures and deterministic providers. They do not establish a production heap limit or affected installation count. Rate-limit reads remain uncached. ## Model Used - Original listing optimization: Anthropic Claude Opus 5.5 (claude-opus-5-5), Claude Code, extended thinking and tool use, as recorded by the original author. - Follow-up fixes and verification: OpenAI Codex, GPT-6, with shell execution, file edits, database fixtures, and browser tests. The exact serving model ID and context window 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 - [x] My branch name describes the change 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: Claude Opus 5.5 <noreply@anthropic.com> Co-authored-by: Dotta <bippadotta@protonmail.com> Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-02 07:45:27 -05:00
Devin FoleyandPaperclip	4a999089ef	Retry transient failures in dashboard reads (#14873) ... ## Thinking Path > - Paperclip shows company activity in the dashboard. > - The dashboard reads company, task, approval, and cost data. > - A pooled database connection can close during one of these reads. > - The driver must reject ambiguous statements because writes may have committed. > - These four dashboard queries are known to be read-only. > - This change retries only the failed read and preserves completed work. ## Linked Issues or Issue Description Refs #14773, which correctly removed automatic replay of ambiguous database statements. Searched existing database and dashboard PRs. Related #8780 changes pool recycling and logging; #13925 adds dashboard consistency coverage. Neither provides these per-query retries. **What happened?** A dashboard request returns a server error when its company lookup, task count, approval count, or monthly spend query loses its database connection. Drizzle wraps the driver's connection error in `cause`. **Expected behavior** A transient connection failure gets a bounded retry of the specific read. Completed reads and budget processing are not replayed. Persistent outages and non-connection errors still fail the request. **Steps to reproduce** Inject a typed `CONNECTION_CLOSED` error into one of these reads. Then allow the next query to succeed. Before this change, the dashboard request fails immediately. ## What Changed - Apply independent retries to the company lookup, task counts, pending approval count, and monthly spend query. Each callback rebuilds its own query with the same company scope. - Extract the existing authentication retry helper as `retryIdempotentDatabaseOperation`. Preserve authentication behavior and its existing exports. - Retain the existing limit of three total attempts with 50 ms and 100 ms pauses. Match typed connection codes through the error cause chain. - Test later-read failures, unchanged query parameters, retry exhaustion, missing companies, and errors that must not retry. Document the boundary. ## Verification - Final focused dashboard, authentication, and real database wire suites: 38 tests passed. Six initial recovery regressions failed before the implementation. - `pnpm -r typecheck`: passed on the final source. - Independent review: no actionable findings. The reviewer separately passed all 38 focused tests and checked the code allowlist, attempt bounds, pauses, and final error identity. - `pnpm test:run`: the general-server group completed with 14,738 tests passed, 13 failed, and 87 skipped. All 13 failures match the previously reproduced clean-base macOS skill-cache failures. The two test files and their implementations are unchanged from that baseline. The runner exited after this group, so the remaining local workspace and serialized groups did not run. All corresponding Linux CI groups passed on this commit. - `pnpm build`: passed on the final source. - Full CI: 53 successful checks and 2 skips on `6a113529c0`. Greptile: 5/5 on that commit, with no review threads or remaining findings. - Merge compatibility with master `f2e0f19630`, including #14866: no conflicts. The five reviewed files are unchanged in the resulting merge tree. ## Risks A persistent outage adds at most two retries per covered query. Each connection attempt retains the configured driver timeout. The change does not repair the underlying network or database failure. Agent counts, run-activity queries, and the budget workflow stay outside these retry boundaries. General database statements and disconnected transactions are not replayed. There is no schema or authorization change. ## Model Used OpenAI GPT-6 through Codex, with reasoning, repository inspection, code editing, and test execution. The runtime does not expose a more specific serving model 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 and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass (38 focused tests; the full local run has the baseline limitation documented above) - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-01 19:23:47 -07:00
Devin FoleyandPaperclip	f2e0f19630	Defer agent directory cleanup until stop proof is available (#14866) ... ## Thinking Path > - Paperclip manages agents and their persistent files. > - Each run owns a temporary agent directory and a save receipt. > - Cleanup needs independent proof that the owning process stopped. > - A cleanup call without that proof currently waits for the directory lock anyway. > - A second lock failure can prevent environment release after the run already reported a failed save. > - This change skips cleanup that has no authority and retries unavailable remote copies after exact destruction proof. > - The save failure stays visible. Existing lock owners remain protected. ## Linked Issues or Issue Description Related work: Refs #14787 (lock diagnostics), #14695 (warm instruction ownership), #9667 (stale lock proposal), and #9872 (control-plane ownership proposal). I checked open PRs and issues. This change leaves the shared filesystem lock protocol in place and does not duplicate the warm-retention work in #14695. **What happened?** Heartbeat cleanup records an explicit unavailable instruction-save warning, then calls directory release before releasing the environment lease. Release can wait for a lock even though the copy has no process-stop proof and cannot be removed. That secondary timeout prevents the following lease-release step. If destruction proof arrives later, the unavailable copy is excluded from both recovery queries. **Expected behavior** Skip a release that cannot remove anything. Preserve the failed-save receipt and candidate fields. Once exact remote destruction is recorded, recover remote cleanup without running a provider command. Unavailable local copies retain their potentially uncollected edits even if local stop proof arrives later. A blocked cleanup must not prevent cleanup for other agents. **Steps to reproduce** 1. Prepare an agent directory, report its save unavailable, and leave process-stop proof absent. 2. Hold the shared directory lock and call release. Before this change, release waits and fails although removal is not authorized. 3. Record destruction of the copy's exact remote lease. Before this change, neither recovery sweep selects the unavailable copy. **Paperclip version or commit** Reproduced against `efc2e6810e9bc0dc8cb412b0e7647c0db9821caa`. **Deployment mode** Local and remote execution with persistent agent directories. Tests use an isolated embedded PostgreSQL database and fixture transports. ## What Changed - Re-read receipts and skip release before lock acquisition when stop proof is absent, the copy is superseded, or cleanup is complete. Keep the same checks inside the lock. - Recover unavailable remote copies only after exact destruction proof. Preserve their unavailable state, errors, candidate hash, and candidate bytes. Keep unavailable local copies and their uncollected edits unchanged. - Store destruction-only cleanup authority with the stop proof. Later cleanup honors it after a lost database response or restart, including when a transport remains cached. - Defer failed or unproven cleanup with bounded batches and a retry delay. Keep failed cleanup visible in logs and its receipt. - Serialize preparation of an existing run with cleanup. Fresh run preparation keeps its existing admission path. - Cover held locks, receipt scope, delayed proof, batch fairness, lost update responses, cached transports, and concurrent same-run preparation with database regressions. ## Verification - Focused directory, legacy instruction-copy, shared lock, and bounded diagnostic suites: 169 tests passed across four files. - `pnpm -r typecheck`: passed on the final source. - `pnpm build`: passed on the final source. - Completed all selected local `pnpm test:run` groups: 733 general server suites, 149 serialized suites, and 14 workspace projects. There are 13 known macOS `EACCES` failures in the unchanged runtime skill cache tests. Their exact signatures match earlier clean-base results, and the cache source and test blobs match both that base and this PR base (existing fix: #14290). One CLI import test timed out under concurrent load; its full file passed separately (17 tests). Broad coverage began before the review corrections; the final source has the focused 169-test run, typecheck, and build. This is a local verification limit, not a passing full local suite. - `git diff --check` and local Gitleaks plus private-identifier/PII diff scans passed. - Independent review of the final source found no remaining actionable issue. Its 17 targeted tests cover crash recovery, cached transports, same-run preparation, real local edit preservation, proof scope, and batch fairness. The main focused run also covers contained scheduling failures. - Final commit `35a24085f7`: Greptile 5/5 with no recommendations and zero unresolved review threads. - Final commit `35a24085f7`: all 53 checks passed, including Canary Dry Run and the security scan; two visual checks were intentionally skipped. The workspace shard passed on retry after GitHub reported that its first runner lost communication. An earlier Canary runner shut down after the release dry run passed. Neither interruption recorded an application assertion failure; the exact final-head checks are now green. ## Risks - This repairs cleanup ordering and recovery eligibility. It does not repair an ambiguous legacy lock owner or restore unsaved files. Actual collection still fails visibly when its lock cannot be acquired. - An unavailable remote copy is recovered only after exact destruction proof. A stopped but retained environment stays protected; recovery does not execute a command that could restart it. - Unavailable local copies with later stop proof still retain potentially uncollected edits. A general local recollection or reclamation policy remains outside this change. - Existing-run preparation now waits for the same lock as cleanup. The fresh-run path is unchanged. - The cleanup mode is stored in the existing private receipt JSON. No schema migration or public API change is required. - No deployment, task replay, or runtime lock deletion was performed. ## Model Used OpenAI GPT-6 (Codex), with reasoning, repository tools, and test execution. The runtime does not expose a more specific model suffix 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 #` 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 - [ ] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>	2026-10-01 14:28:47 -07:00
DottaandPaperclip	4039d4f06b	fix(auth): allow scoped low-trust work and owner-chat instruction edits (#14870) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Low-trust agents must work within their assigned scope. > - Task creation currently rejects these agents before checking assignment permission or scope. > - Persistent instruction saves also reject direct requests from authorized chat owners. > - This PR checks the requested action and its recorded authority instead of denying all such work. > - Agents can organize permitted work and follow their owner's instruction-edit requests while outside work stays restricted. ## Linked Issues or Issue Description **What happened?** Low-trust agents cannot create self-assigned tasks or subtasks, even within their allowed scope. An authorized user also cannot ask an agent in their own Agent Chat to update its managed `AGENTS.md`. Agent-folder collection can hide the permission rejection behind a generic save failure. **Expected behavior** Allow task creation when assignment permissions and project or root-task scope permit it. Allow instruction self-edits during authenticated owner-chat execution, subject to the user's current edit permission. Outside tasks, subtasks, connector messages, and peer agents do not inherit that instruction authority. Explain the actual denial when a save fails. **Steps to reproduce** 1. Configure an active agent with `low_trust_review` and a project or root-task boundary. 2. Ask it to create an in-scope task assigned to itself, or a subtask of its own task. 3. As a user with permission to configure that agent, ask it in your Agent Chat to update its managed `AGENTS.md`. 4. Observe blanket permission denials rather than action-specific checks. **Paperclip version or commit** Rebased onto master at `8ec4b84e1`. This is a core authorization change, independent of adapter choice. **Deployment mode** Authenticated server. Regression coverage uses the server services, HTTP routes, native tool authority, and embedded PostgreSQL. Related work: #14775 adds human-directed task execution. #13599 concerns instruction-path configuration; this PR leaves that configuration restricted. #11988 proposes separate active-review instruction protection. #10693 reports unclear authorization denials on a different API surface. ## What Changed - Apply task-assignment checks to both HTTP creation routes and native task creation, including unassigned work. Preserve low-trust policy and source attribution on the created task and its initial plan. - Allow self-assigned decomposition within the permitted project or root-task tree. Resolve workspace-derived project scope before authorization, and reauthorize existing tasks before duplicate detection returns them. Keep cross-project and peer-assignment checks. - Derive instruction self-edit authority from the accepted run identity and authenticated owner-message wake. Recheck current permissions at save time. Bind retries to the same request and chat session. - Reject inherited instruction authority from outside tasks, subtasks, plugins, connectors, stale sessions, cancelled runs, and peer edits. - Surface permission errors in instruction and agent-folder save receipts. Tell chat agents to explain the rejected action and the specific restriction. - Update the low-trust policy and implementation documentation. ## Verification - All 297 tests in 11 focused server suites pass after the rebase. These cover owner-chat saves, private copies, warm agent directories, reset and retry boundaries, permission revocation, task creation routes, and native tool authority. - After review fixes, all 126 tests in the four affected authorization/chat suites pass. Workspace scope regressions and 146 existing creation/ownership/workspace-route tests also pass. - The final duplicate-task and CI fixes pass all 39 tests across chat-project tools, duplicate creation, environment-selection guards, and assignee-invokability routes. The duplicate-task test reproduced an unauthorized response before the fix and verifies denial plus permitted reuse afterward. - `pnpm --filter @paperclipai/server typecheck` passes after rebasing; `pnpm --filter @paperclipai/server exec tsc --noEmit` also passes after the review fixes. - `git diff --check origin/master...HEAD` passes. - Final head `7e73270b86748792649e4ae6fbc6879f73b42b73`: all 54 checks passed, with two expected skips and no pending or failed checks. This includes builds, typechecking, the full test matrix, end-to-end tests, runner verification, the canary dry run, and security scans. - Greptile is 5/5 on that exact head, with no unresolved review threads. This change has not been deployed to staging. ## Risks This changes authorization behavior. The instruction exception must not become an inherited task permission. The check uses server-owned execution records, requires the agent's own chat and instructions, and keeps normal protected-change and responsible-user checks. Saves fail closed when current provenance or permission is missing. Owner chat grants a turn-scoped capability; the server does not classify the message intent or require approval of the exact new file bytes. Prompt injection within an authorized owner-chat turn remains a model-level risk. This is the requested owner-chat trust boundary, without a new per-edit confirmation flow. No database migration or broad trust-preset change is required. ## Model Used OpenAI Codex, based on GPT-6, with reasoning, code editing, shell tools, and test execution. The exact runtime 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>	2026-10-01 15:42:40 -05:00
DottaandPaperclip	8ec4b84e1c	fix(chat): resume messages after failed runs without duplicate delivery (#14857) ... ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A user can send a new message after a native run fails. > - The server checks that the old execution has stopped before it starts a fresh turn. > - A failed run can retain a result accepted before checkpoint or cleanup failed. > - The continuation gate treated that saved result as active recovery and held the new message forever. > - This pull request removes that false liveness signal while retaining controller, process, environment, and authorization checks. > - Live staging then exposed a second defect: chat admission created a successor without consuming the original deferred receipt, so completion delivered the message again. > - Consume that exact receipt atomically with admission, while preserving separate turns for later chat messages. ## Linked Issues or Issue Description **What happened?** A new user message stayed in the queue with `controller_settling` after the previous run had reached `terminal_failure`. The old coordinator had no lease owner but still had a `resultId`. Its remote environment had a verified stop receipt. **Expected behavior** Start one fresh turn after execution has stopped and normal admission checks pass. Preserve the failed run and its accepted result as history. **Steps to reproduce** 1. Accept a native result, then fail checkpoint or cleanup and exhaust recovery. 2. Retain the result ID on the terminal failure record and stop the execution environment. 3. Send a new user message. Before this fix, it waits forever for the finished controller. **Paperclip version or commit** Reproduced in a database-backed regression test on `26900655b`. **Deployment mode** Server with a native runner and remote sandbox. Local process stop checks also apply. Related: https://github.com/paperclipai/paperclip/pull/14775. Searched existing PRs for retained-result continuation fixes; no duplicate found. ## What Changed - Remove the retained-result veto for terminal failures. - Keep controller ownership, successor, process, environment cleanup, pending decision, and ordinary admission checks. - Add regressions for retained results, active execution, missing stop evidence, and delayed remote cleanup. - Atomically consume the resumed receipt in agent chat, even though chat does not coalesce other queued messages. - Reproduce completion-time duplicate promotion, race cleanup against periodic recovery, and prove a subsequent chat message keeps its own turn. - Document that a saved result does not make a terminal failure active. - Keep exhausted workspace export on its separate repair path, tested through the production finalizer. ## Verification - Red: retained-result admission failed with `controller_settling` before the original fix. The new chat-specific regression then reproduced duplicate promotion when the first reply finished. - Green: 406 tests across native continuation, workspace-export recovery, and the wake-queue module passed on `cbc531cc0`. - The chat regressions exercise real Postgres transactions, simultaneous recovery callbacks, successful completion, the production queue-drain use case, and repeated drain attempts. A distinct follow-up remains a separate turn. - `pnpm -r typecheck` and `pnpm build` passed on `cbc531cc0`. - The earlier full local test run encountered a timeout and follow-on failure in unchanged AI connection-adoption tests; all 50 tests passed on isolated rerun. That local run was stopped after the full CI test matrix passed on the earlier head. - All 54 CI checks passed on `cbc531cc0` (2 skipped), including the full test matrix and browser shards. One unchanged interaction-route test returned HTTP 500 on its first CI attempt; its full 84-test file passed locally, and the failed shard passed on one targeted rerun. - Greptile reviewed `cbc531cc0`: 5/5, no unresolved findings. - Live staging first verified that the original saved message resumes and receives a successful response; that test exposed the duplicate now covered above. - Deployed exact commit `cbc531cc0410e1ef6e8811c6c5c014c3528351ed` to the affected staging workspace; deployment verification, health, authentication, and startup recovery passed. - Submitted a fresh message through the browser. The agent replied in 39 seconds; server records show exactly one successful run, native phase `committed`, no error, and an empty queue. A later check more than a minute after completion found no duplicate run. ## Risks The change affects admission after native execution failure and consumption of a resumed deferred receipt. A fresh turn must never overlap the prior execution, and consuming one chat receipt must not absorb later messages. Tests retain the controller, process, and remote-stop guards. This change does not migrate data, apply an old result, or reset the old retry budget. ## Model Used OpenAI Codex (GPT-6). The exact runtime model identifier and context window are not exposed in this session. Used reasoning, repository inspection, code execution, database-backed tests, and browser inspection. ## 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>	2026-10-01 14:43:02 -05:00