mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-10-17 10:06:42 +02:00
95026970221bc72041330c8542abe31e2f80ca4b
4938
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 DottaandPaperclip
|
9502697022 |
Require authenticated Copilot model verification before agent adoption
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
5cb6ad5207 |
Carry the maintained master prerequisite through the Copilot review stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '6ba290a8a271772907917695a7c77d414989c23a': fix(ui): add icons to chat connection choices (#15574) Pin Copilot native dependencies with the maintained lockfile refresh (#15572) refactor(server): extract heartbeat run-log formatting (#15573) refactor(server): extract heartbeat task markdown rendering (#15568) test(evals): verify durable waiting across continuation checkpoints (#15564) fix(slack): upload agent avatars directly during Cloud setup (#15566) |
||
|
DottaandPaperclip
|
6ba290a8a2 |
Carry the maintained master prerequisite through the Copilot review stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit 'f6d04cd13502b5231a763e9c447f11aa95c67cca': fix(ui): add icons to chat connection choices (#15574) Pin Copilot native dependencies with the maintained lockfile refresh (#15572) refactor(server): extract heartbeat run-log formatting (#15573) refactor(server): extract heartbeat task markdown rendering (#15568) test(evals): verify durable waiting across continuation checkpoints (#15564) fix(slack): upload agent avatars directly during Cloud setup (#15566) |
||
|
DottaandPaperclip
|
f6d04cd135 |
Carry the maintained master prerequisite through the Copilot review stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '33f7e7b4644f4fc386f44f17ac04846abf062f7f': fix(ui): add icons to chat connection choices (#15574) Pin Copilot native dependencies with the maintained lockfile refresh (#15572) refactor(server): extract heartbeat run-log formatting (#15573) refactor(server): extract heartbeat task markdown rendering (#15568) test(evals): verify durable waiting across continuation checkpoints (#15564) fix(slack): upload agent avatars directly during Cloud setup (#15566) |
||
|
DottaandPaperclip
|
33f7e7b464 |
Integrate maintained master and its CI-owned Copilot dependency prerequisite
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit 'd6df12cef69fcaf2d2fe66a393168931d5b8b4e7': fix(ui): add icons to chat connection choices (#15574) Pin Copilot native dependencies with the maintained lockfile refresh (#15572) refactor(server): extract heartbeat run-log formatting (#15573) refactor(server): extract heartbeat task markdown rendering (#15568) test(evals): verify durable waiting across continuation checkpoints (#15564) fix(slack): upload agent avatars directly during Cloud setup (#15566) |
||
|
DottaandPaperclip
|
fe1ef2ba36 |
Enforce Copilot early finish and close final qualification review gaps
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
d6df12cef6 |
fix(ui): add icons to chat connection choices (#15574)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Connectors let people chat with agents and let agents use external tools. > - Slack setup asks the user to choose between those two uses. > - Both choices currently use text alone. > - This pull request adds an agent avatar and a hammer to the shared chooser. > - The icons help users identify each choice at a glance. ## Linked Issues or Issue Description **What existing behavior does this improve?** The chat-or-tools choice in connector setup. Related setup work: Refs #15413. **Subsystem affected** ui/ — React board UI and Storybook. **Current behavior** Both choice cards show a title and description with no identity icon. **Proposed behavior** Show the existing agent avatar beside chat. Show a hammer beside tools. Align both icons and keep the existing navigation. **Reason and benefit** Users can tell the two choices apart with less reading. **Breaking changes** None. The shared chooser applies the icons wherever that choice appears. Only Slack currently offers both paths in the catalog. Other chat-only providers keep their direct setup path. ## What Changed - Add the shared agent avatar to the chat choice. - Add a decorative hammer icon to the tools choice. - Add desktop and mobile Storybook stories that render the production page. ## Verification - Token gates pass. - Focused catalog, routing, and UI contract tests pass, including the new chooser accessibility and navigation test. - Checked the production chooser in Storybook at desktop and mobile widths. - Local repository typecheck, build, and Storybook build pass. - All 56 checks pass on commit `0f0673f1bb5cd340084f3a5229983da049f1c8db`, including the full CI test suite and browser shards. Greptile reports 5/5 with no findings. - Started the full local test command. Stopped that duplicate run after the full CI suite passed. The focused local tests completed successfully. - In Storybook, open Connections → Slack → Automatic setup → 00 · Choose connection. The mobile variant is in the same group. ## Risks Low risk. This changes card layout and decorative images. Existing button text, click handlers, credentials, and permissions stay the same. The avatar is generic because the user has not selected an agent yet. ## Model Used OpenAI GPT-6 through Codex. Used code inspection, edits, shell tools, and browser tools. The session does not expose the exact runtime 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> |
||
|
|
941a3fa991 |
Pin Copilot native dependencies with the maintained lockfile refresh (#15572)
Pin the three optional GitHub Copilot 1.0.88 native packages for Runner and server using the maintained lockfile workflow. Synchronize the package contract and bound initial render readiness in the deliberately throttled browser fixture. Current-head CI and focused checks pass. Co-Authored-By: Dotta <cryppadotta@users.noreply.github.com> Co-Authored-By: lockfile-bot <lockfile-bot@users.noreply.github.com> Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
87a7312cfd |
refactor(server): extract heartbeat run-log formatting (#15573)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Heartbeat orchestration records agent output and run events. > - The heartbeat service still contains more than 30,000 lines after the first extraction. > - Log formatting has fixed limits and does not write run state. > - These helpers form a small next step toward a more manageable heartbeat service. > - This PR moves them into the heartbeat folder without changing their function bodies. > - Direct boundary tests and existing caller tests check the output. ## Linked Issues or Issue Description Refs: #15568. **What existing behavior does this improve?** This improves the structure and test coverage of heartbeat run-log formatting. **Subsystem affected** server/ — orchestration services. **Current behavior** `heartbeat.ts` contains excerpt handling, payload size limits, and log chunk formatting beside run orchestration. **Proposed behavior** Move these helpers and their constants into `server/src/services/heartbeat/run-log.ts`. Keep the same function bodies, output, limits, and public exports. **Reason and benefit** Run-log formatting has its own small module and focused tests. This keeps the second extraction small enough to review on its own. **Breaking changes** None. The existing public import path and persisted output stay the same. Related run-log PRs: Refs: #6373, Refs: #8841. Those PRs change redaction behavior. This PR only moves existing formatting code. ## What Changed - Move 109 lines of helper functions and six constants into `heartbeat/run-log.ts`. - Move `appendExcerpt` and retain the existing public exports for `boundHeartbeatRunEventPayloadForStorage` and `compactRunLogChunk` in `heartbeat.ts`. - Add 17 direct regression cases for size and depth limits, cycles, shared references, immutable input, image omission, redaction order, excerpt tails, and UTF-8 boundaries. - Document both heartbeat extractions in `doc/DEVELOPING.md`. ## Verification - Before extraction, four focused files passed with 55 tests. - After extraction and the two new excerpt tests, the same four files passed with 57 tests. - Run `pnpm exec vitest run server/src/services/heartbeat/run-log.test.ts server/src/__tests__/heartbeat-run-log.test.ts server/src/__tests__/heartbeat-list.test.ts server/src/__tests__/redaction.test.ts`. - The original and extracted function blocks match byte for byte after adding the export keyword to `appendExcerpt`. - Existing tests still import the public helpers from `heartbeat.ts`. - `node scripts/check-module-boundaries.mjs` and `git diff --check` passed. - `pnpm -r typecheck` passed. - `pnpm build` passed. - The full suite passed in CI. The duplicate local `pnpm test:run` was stopped after CI passed; it did not complete locally. - All CI gates passed on commit `4e65c666e3b5b46162b3a528b37fba8902a1d200`: 54 passing checks, two skipped Storybook checks. - Fresh Greptile review of the same commit: 5/5, no actionable or inline findings. The PR has no merge conflicts. ## Risks - Low risk. Moving code can cause an import or build error. - The same redaction and adapter utility modules remain in use. - Database writes, current-user redaction, live event delivery, and run state remain in `heartbeat.ts`. - The event schema and payload output do not change. - No database, API, Telemetry, Observability, or UI contract changes are required. > For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and discuss it in `#dev` before opening the PR. Feature PRs that overlap with planned core work may need to be redirected — check the roadmap first. See `CONTRIBUTING.md`. ## Model Used OpenAI GPT-6 via Codex. The exact serving model ID and context window were not exposed in this session. Used reasoning, repository inspection, shell tools, 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> |
||
|
DottaandPaperclip
|
fc80f1d100 |
Carry the explicit Copilot onboarding fixture policy into qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '3eeb21f50a13c3ea2c0c65778f315ce7d747b1e6': Make positive Copilot onboarding fixtures declare native Runner readiness |
||
|
DottaandPaperclip
|
3eeb21f50a |
Make positive Copilot onboarding fixtures declare native Runner readiness
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
dcf671192b |
Make positive Copilot onboarding fixtures declare native Runner readiness
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
51653becc2 |
refactor(server): extract heartbeat task markdown rendering (#15568)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Heartbeat orchestration supplies each run with task context. > - The task Markdown renderer is inside a service with more than 31,000 lines. > - The renderer formats task data and does not write run state. > - It is a small first step toward a more manageable heartbeat service. > - This PR moves the renderer without changing its function body or public export. > - Direct tests and existing caller tests check the output before and after the move. ## Linked Issues or Issue Description **What existing behavior does this improve?** This improves the structure and test coverage of heartbeat task context rendering. **Subsystem affected** server/ — orchestration services. **Current behavior** `buildPaperclipTaskMarkdown` occupies 352 lines inside `heartbeat.ts`. The service also owns run execution, scheduling, recovery, and cancellation. **Proposed behavior** Move the renderer to `heartbeat/task-markdown.ts`. Keep its function body and the existing `heartbeat.ts` export unchanged. Leave orchestration in place for this first PR. **Reason and benefit** Prompt formatting has its own small source file and direct regression tests. Reviewers can check this one extraction before a later refactor. **Breaking changes** None. The input type, rendered output, and existing import path stay the same. Related renderer changes: Refs: #4732, Refs: #14030. These PRs change prompt behavior. This PR only moves the current renderer. ## What Changed - Move the 352-line renderer to `server/src/services/heartbeat/task-markdown.ts`. - Import and re-export it from `heartbeat.ts`. Remove imports used only by the renderer. - Add seven direct regression cases. Cover empty context, ordered comment-only wakes, input preservation, nested code fences, ancestor limits, attachment-only wakes, and rejected plans. - Keep the renderer and its direct tests in `server/src/services/heartbeat/`. Document this folder as the home for relevant later extractions in `doc/DEVELOPING.md`. ## Verification - The focused suite passed before and after extraction, and after moving to the heartbeat folder: four files, 85 tests. - Run `pnpm exec vitest run server/src/services/heartbeat/task-markdown.test.ts server/src/__tests__/heartbeat-context-summary.test.ts server/src/__tests__/heartbeat-chat-task-link.test.ts server/src/__tests__/codex-local-execute.test.ts`. - The extracted function matches the original function byte for byte. The folder move only adjusts imports. - `node scripts/check-module-boundaries.mjs` passed. - `git diff --check` passed. - The initial extraction passed local `pnpm -r typecheck` and `pnpm build`. - The folder update passed `pnpm --filter @paperclipai/server exec tsc --noEmit` and `pnpm --filter @paperclipai/server build`. - The initial extraction passed the full CI suite. Its duplicate local `pnpm test:run` was stopped after CI passed; it did not complete locally. - The folder update passed all CI gates on commit `e21e589b379d7ca2fae16c1dcd91cf2f604fd03f`: 54 passing checks, two skipped Storybook checks. Three test shards passed after one retry following simultaneous runner shutdowns; those failures had no failed test assertions. - Fresh Greptile review of commit `e21e589b379d7ca2fae16c1dcd91cf2f604fd03f`: 5/5, no actionable or inline findings. ## Risks - Low risk. A moved module can change import resolution or expose an import cycle. - Existing caller tests still load the compatibility export from `heartbeat.ts`. - The same guidance constants and public task URL resolver remain in use. - No run-state writes, transactions, locks, shared process state, or cleanup paths moved. - No database, API, telemetry, or UI contracts changed. > For core feature work, check [`ROADMAP.md`](ROADMAP.md) first and discuss it in `#dev` before opening the PR. Feature PRs that overlap with planned core work may need to be redirected — check the roadmap first. See `CONTRIBUTING.md`. ## Model Used OpenAI GPT-6 via Codex. The exact serving model ID and context window were not exposed in this session. Used reasoning, repository inspection, shell tools, 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>canary/v2026.1008.0-canary.11 |
||
|
DottaandPaperclip
|
d99959f00b |
Carry the verified Copilot setup review fixes into qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '75b5379a6bcb63a584406421154b091084edfe87': Refresh first-token Copilot discovery and honor native Runner admission in onboarding |
||
|
DottaandPaperclip
|
75b5379a6b |
Refresh first-token Copilot discovery and honor native Runner admission in onboarding
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
7010b852e4 |
Refresh first-token Copilot discovery and honor native Runner admission in onboarding
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
24a15c5afa |
test(evals): verify durable waiting across continuation checkpoints (#15564)
## Thinking Path > - Paperclip manages agents and durable tasks across provider runs. > - Human questions must preserve task ownership and stop work until a real answer arrives. > - The continuation eval checks the final work, but its lifecycle oracle misses several broken waiting states. > - A correct final answer can hide a stale execution lock or a lost intermediate answer receipt. > - This pull request checks each wait and retains every question and run identity. > - A source audit records which waiting operations belong to native runtimes and which still require legacy API calls. ## Linked Issues or Issue Description **What existing behavior does this improve?** The existing Product E2E continuation oracle for human question and approval journeys. **Current behavior** A pending interaction can pass the waiting check even when its task has the wrong status, a stale lock, or a retry. Only the first pending question and first run receipts are checked at the final checkpoint. **Proposed behavior** Require a healthy wait on the same task and assignee. Preserve all intermediate run receipts and every pending question's answered identity. Allow a paused native provider question only when its pending runtime request identifies the running native run and execution lock. Related: #15548 and #15554 cover earlier bookkeeping slices. #15544 changes production continuation summaries; this PR changes the eval oracle and does not overlap that fix. ## What Changed - Check waiting state, execution ownership, and all answer/run identities in the existing lifecycle oracle. - Add negative calibrations for broken records and positive coverage for both semantic waits and paused provider questions. - Retain task activity at each checkpoint for inspection of successful persisted mutations. - Verify 1,000-cent company and agent budget limits before continuation work. Disable automatic cell rerolls. - Document runtime ownership, existing coverage, and the remaining instruction decision. Keep production instructions unchanged. ## Verification - `pnpm test:e2e:runner:unit`: 1,867 Vitest tests pass, one skips; 128 Node tests pass. - `pnpm test:e2e:runner:typecheck`: passes. - `pnpm -r typecheck`: passes. - `pnpm build`: passes. - Negative calibration: 27 added cases fail against the prior oracle and pass with these checks. - Four existing local continuation cells are selected for a separate bounded live canary. Live results are pending; no behavioral pass is claimed here. - The full local `pnpm test:run` suite was not repeated because embedded PostgreSQL was unavailable in the preceding workspace verification. Required Linux PR CI must pass before readiness. ## Risks The stronger oracle can expose existing product or fixture defects. A paused provider question and a terminal semantic wait have distinct valid states. The four-cell canary does not qualify approval/review, dependency unblock, crash races, remote execution, or general task quality. Task activity records successful persisted writes, not failed API attempts; repeated progress comments are not automatically defects. Historical eval grades remain unchanged. No production scheduling, prompt, tool, schema or migration changes. ## Model Used OpenAI Codex, GPT-6-based, with tool use and code execution. The exact serving model ID and context-window size are not exposed in this session. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [ ] 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> |
||
|
DottaandPaperclip
|
1620e373b6 |
Synchronize the provider-death fixture with its shared lifetime quorum
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
a179708f4f |
Carry the verified provider fixture synchronization through Copilot setup
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '1620e373b6602de079c8637951273b22bf86c2c3': Synchronize the provider-death fixture with its shared lifetime quorum |
||
|
DottaandPaperclip
|
a7212bd765 |
Carry final Copilot setup and fixture review repairs into qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit 'a179708f4f1b27e91830672fcd35c46669e0266c': Offer Copilot in ordinary New Agent creation Synchronize the provider-death fixture with its shared lifetime quorum |
||
|
DottaandPaperclip
|
d2c65784cb |
Offer Copilot in ordinary New Agent creation
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
05daf995d1 |
Preserve exact remote bootstrap read evidence in provider-death qualification
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
5e790deff5 |
Offer Copilot in ordinary New Agent creation
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
cf37914986 |
Synchronize the provider-death fixture with its shared lifetime quorum
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
f47614046d |
fix(slack): upload agent avatars directly during Cloud setup (#15566)
## Thinking Path > - Paperclip helps people manage AI agents for work. > - Slack setup creates a dedicated bot for an agent. > - The setup should upload that agent's avatar. > - The current upload asks Slack to fetch an image from the board origin. > - Cloud requires a tenant session at that origin, so Slack receives HTTP 401. > - This change renders the PNG on the server and uploads the file directly. ## Linked Issues or Issue Description Refs #15413. **What happened?** Automatic Slack setup left the app and bot with default icons on Cloud staging. An unauthenticated request to the exact avatar URL returned HTTP 401 with `tenant_session_required`. Local setup did not expose this Cloud ingress requirement. **Expected behavior** New Slack bots receive the assigned agent's 512-pixel avatar with the Paperclip dark background. **Steps to reproduce** 1. Create a Slack app through automatic setup on a Cloud tenant. 2. Complete installation. 3. Inspect the bot avatar in Slack. The previous URL-based upload cannot fetch the image without a tenant session. ## What Changed - Render the assigned agent's preset PNG with the existing bounded worker pool. - Send PNG bytes as multipart `file` data to `apps.icon.set` instead of passing a board URL. - Keep the temporary token in the Authorization header. Let fetch set the multipart boundary. - Recheck management permission and credential-lease ownership after rendering. Close the worker pool during chat service shutdown. - Cover actual PNG dimensions, uploaded bytes, failure recovery, and secret-safe responses. Update deployment documentation. ## Verification - Passed: 84 focused tests across automatic Slack registration and on-demand agent avatars. - Passed: full repository build. - Passed: full repository typecheck. All 54 current-head GitHub checks passed, including the complete test matrix, all browser shards, build, typecheck, canary, and security checks. Greptile completed on `79e79f6fc` with 5/5 and no actionable findings or open review threads. - The Cloud fetch failure was reproduced without browser credentials. No Cloud access rule was changed. - A real Slack upload with this new path still requires deployment and a fresh automatic setup. Existing apps retain the manual avatar-upload fallback. ## Risks - The renderer can time out or Slack can reject the upload. Both failures preserve the saved app and leave installation usable. - The renderer adds a bounded, lazy worker pool to Slack registration. Shutdown closes it. - No migration, bot permissions, credential retention, or Cloud authentication behavior changes. ## Model Used OpenAI Codex, GPT-6 family, with reasoning, repository tools, code execution, and browser inspection. The runtime does not expose a more specific authoring 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>canary/v2026.1008.0-canary.10 |
||
|
DottaandPaperclip
|
2f390c1710 |
Carry focused Copilot review repairs through the stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '6170202dacb9c299e79f2dd45e09b419e19839a4': Always remove private Copilot probe roots after cleanup or report failure |
||
|
DottaandPaperclip
|
4acb084fc5 |
Carry focused Copilot review repairs through the stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit '2f390c171086fb27ad6871a5d64b8aa205d650d7': Preserve repository credentials and Copilot authentication failures Always remove private Copilot probe roots after cleanup or report failure |
||
|
DottaandPaperclip
|
32b9509390 |
Preserve repository credentials and Copilot authentication failures
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
e489a2825a |
Always remove private Copilot probe roots after cleanup or report failure
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
6170202dac |
Carry focused Copilot review repairs through the stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> * commit 'e489a2825ae47461ef265b0cc90b40f19e634a8f': Always remove private Copilot probe roots after cleanup or report failure |
||
|
DottaandPaperclip
|
266035f210 |
Reject extra provider-death operations and lost process identity
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
145f3f1725 |
Preserve repository credentials and Copilot authentication failures
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
6582e7057a |
Always remove private Copilot probe roots after cleanup or report failure
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
cc8ad29063 |
Carry the reviewed clean-install fixture repair through the Copilot stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
f2ac27a9f7 |
Carry the reviewed clean-install fixture repair through the Copilot stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
7075a573bd |
Keep Copilot installer contracts complete in the runtime partition
Include the server lifecycle shim with its package tests. Retain extended native-session coverage in the following regression partition so each review stays below 100 files. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
dbbe3964fc |
Carry the reviewed clean-install fixture repair through the Copilot stack
Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
58a7cebac3 |
Compile real Copilot installation fixtures in clean checkouts
Remove dependence on a prebuilt workspace dist while retaining installation, closure-tampering, foreign-link, and bootstrap assertions. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
e199b349f6 |
Compile real Copilot installation fixtures in clean checkouts
Remove dependence on a prebuilt workspace dist while retaining installation, closure-tampering, foreign-link, and bootstrap assertions. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
fcbb80554e |
Add saved-token Copilot setup and normal runtime admission
Reviewed partition of the current-master Copilot integration. Complete v30 live qualification is retained; current v31 recovery compatibility, CI and review are tracked explicitly. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
c48b89618f |
Add Copilot Product qualification and fixture ownership checks
Reviewed partition of the current-master Copilot integration. Complete v30 live qualification is retained; current v31 recovery compatibility, CI and review are tracked explicitly. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
50368e9c50 |
Add Copilot runtime regressions and provenance fixtures
Reviewed partition of the current-master Copilot integration. Complete v30 live qualification is retained; current v31 recovery compatibility, CI and review are tracked explicitly. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
83841aecf6 |
Add GitHub Copilot 1.0.88 runtime on current master
Reviewed partition of the current-master Copilot integration. Complete v30 live qualification is retained; current v31 recovery compatibility, CI and review are tracked explicitly. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
2f0c485dec |
fix(skills): ship the completion helper with the installed skill (#15554)
## Thinking Path > - Paperclip manages work for AI agents. > - Legacy agents use the Paperclip skill to save task status and comments. > - The skill names a script relative to the task workspace. > - That script exists only in the Paperclip source repository. > - Agents in other workspaces can hit a missing command or search for it. > - This PR ships the helper inside the skill and uses the installed skill path. > - The repository command remains available through a forwarding wrapper. ## Linked Issues or Issue Description Fixes #9527. Refs #15548 for the preceding runtime checkout guidance. Related: #6052 addresses LF line endings for the repository helper; this change addresses helper delivery and path resolution. ## What Changed - Bundle the existing issue update helper with the Paperclip skill. Preserve its HTTP checks, echoed-status check and two-attempt limit. - Resolve the command from the installed skill directory. Use a verified PATCH when that path is unavailable, without searching the filesystem. - Keep the repository command as a wrapper that works from any directory. - Test shell execution and exact status/comment payloads through both provider skill-home layouts, including paths with spaces. - Add helper sources and existing verification tests to stock-harness admission. Record an absent historical helper explicitly. Add the missing declaration for the admission fingerprint export. ## Verification - Complete directly affected source suites: 30 tests pass. They cover skill delivery, preserved multiline comments and links, authentication headers, empty responses, mismatched status, transient retries and definitive rejections. - Product E2E typecheck passes. Support suites: 1,835 Vitest tests pass, one is skipped; 128 Node tests pass. - Full local build and workspace typecheck pass. - Full local repository tests are not claimed as passed. Embedded PostgreSQL was unavailable in this worktree during the preceding task; Linux CI will run the repository gates. - The authorized matched Codex/Claude comparison is pending. It uses the existing assigned-skill case and original oracle, one initial attempt per profile and variant. - CI and a fresh Greptile review are pending. Keep this PR in draft until readiness gates complete. ## Risks - Correct path resolution depends on the harness supplying the installed skill path. The instructions use verified PATCH when that path is unavailable. - The helper still requires Bash, curl and jq. Its existing retry and response-verification behavior is unchanged. - Tests use the shared skill-directory symlink mechanism and an HTTP fixture. Real provider completion behavior still requires the bounded live comparison. - This fix does not redesign native completion, legacy recovery or ambiguous transport handling. ## Model Used OpenAI Codex, GPT-6 family. The exact model build and context window are not exposed in this session. Used code editing, shell tools 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 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>canary/v2026.1008.0-canary.9 |
||
|
DottaandPaperclip
|
71cd0a2621 |
fix(skills): honor the current run harness checkout (#15548)
## Thinking Path > - Paperclip manages work for AI agents. > - The runtime claims eligible assigned tasks before it starts an agent. > - The wake tells the agent when the runtime already holds that claim. > - The legacy skill still requires another checkout in every case. > - This PR makes the skill honor the current task and run claim. > - Manual checkout and server ownership checks remain in place for other cases. ## Linked Issues or Issue Description **Where is the issue?** `skills/paperclip/SKILL.md`, in the scoped wake procedure and Step 5. **What's wrong?** The wake can say that the harness already checked out the issue. The skill still tells the agent that it must call checkout. These instructions conflict. **Suggested fix** Skip the second checkout only when the runtime wake explicitly confirms the claim for this issue and run. Retain manual checkout when that statement is absent or the agent selects another task. Refs #14948 for the existing shared prompt reduction. ## What Changed - Honor the explicit runtime claim in the scoped wake procedure and Step 5. - Keep context reads, status writes, deliverable handling and conflict rules. - Add checks for normal and resumed wake text and excluded automatic claims. - Retain successful checkout HTTP activity for legacy stock-task evals. Bind each receipt to the exact company, task, agent and run. Keep this observation separate from the original task grades. ## Verification - Checkout observation calibration: nine tests pass. - Focused skill, wake and database ownership tests: in progress. - Full repository build, typecheck and tests: in progress. - Planned live comparison: the existing assigned-skill document case on legacy Codex and Claude. One attempt per variant and profile. No automatic retries. The baseline and candidate share the observation code and task oracle. - Live results are pending. This draft does not claim behavioral qualification. ## Risks - Agents may misread prompt guidance. The API still enforces ownership; the text grants no new authority. - The exception is specific to the current issue and run. It does not remove ordinary legacy completion writes or authorize another task. - Activity measures successful checkout HTTP calls. Failed attempts require separate run-log inspection. Missing or mismatched observations cannot count as zero calls. - One trial per profile cannot establish general reliability, speed or cost trends. ## Model Used OpenAI Codex, GPT-6 family. The exact model build and context window are not exposed in this session. Used code editing, shell tools 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 - [ ] 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>canary/v2026.1008.0-canary.8 |
||
|
DottaandPaperclip
|
d66acb7ac1 |
feat: automate Slack bot app setup and installation (#15413)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Chat connectors give each agent a customer-owned bot and task-backed conversations. > - Manual Slack setup requires app creation and copying durable credentials. > - Operators need a shorter setup that an assisting agent can use safely. > - This pull request creates the app through Slack's Manifest API and installs it through OAuth. > - Durable registration state supports recovery without creating another app. > - A four-screen wizard, automatic avatar upload, and OAuth account linking reduce setup work. > - Connector settings and per-turn tool guidance support daily use after installation. ## Linked Issues or Issue Description **Subsystem affected** Native Slack bot setup, company secret storage, chat connector management, and agent tool guidance. **Problem or motivation** New Slack bots require manual app creation and copying a signing secret and bot token. Interrupted setup can create duplicate apps. The setup and management screens contain unnecessary controls. Agents also need guidance for native questions, files, thread replies, and governed Slack actions. **Proposed solution** Use a temporary app-configuration access token to create a customer-owned app. Save durable secrets in the vault. Bind OAuth to the initiating actor, company, endpoint, registration revision, scopes, and configured origins. Preserve manual and existing-app recovery. Link the installing user's account, send a welcome DM, and advance from saved server evidence. Keep request-URL recovery instructions available if automatic connection detection waits. **Alternatives considered** The Slack CLI adds installation requirements. Socket Mode changes transport. A shared Paperclip-owned app changes app ownership. These alternatives are outside this change. **Roadmap alignment** This extends existing chat connectors and secrets capabilities. Related public work: #14037 and #13954 cover Slack MCP prerequisites and user OAuth. No duplicate bot-registration PR was found. ## What Changed - Share one reviewed manifest builder between automatic registration and manual setup. - Add replay-safe migration 0318 and company-bound registration state with vault references and uncertain-creation recovery. - Add registration, installation, callback, and resume APIs with short-lived, single-use OAuth state. - Save installation credentials before downstream checks and preserve bot identity constraints. - Reduce automatic setup to four screens. Keep advanced app details, manual recovery, and existing-app setup. - Upload the agent avatar with the Paperclip dark background. Link the OAuth installer's account and send setup DMs. - Show agent and connector-owner avatars. Simplify settings, access, and conversation screens. - Discover joined Slack channels and enable them by default. Start a task from a bare mention and admit same-thread follow-ups. - Refresh Slack tool guidance each turn. Add native-form, file, approval, and delivery regressions plus manual model probe definitions and sanitized acceptance records. - Update deployment/database docs, OpenAPI, redaction, removal cleanup, production Storybook stories, and provider browser tests. - Merge current master and move the registration migration after its latest migration without rewriting published commits. The completed Slack success view intentionally has a single centered **Done** action and no **Save & exit**, as explicitly requested by the product owner. `DESIGN.md` records this exception; unfinished setup steps retain the aligned wizard footer. ## Verification - Passed after the master merge: repository typecheck, full build, Storybook build, design-token gates, module-boundary gates, and migration generation. - Passed: all 352 focused Slack deterministic tests and all 14 affected provider browser tests. Browser tests use controlled provider fixtures and a separate throwaway instance. - Passed on current head `c5d01e0e2`: the complete GitHub test matrix (general server, chat, all workspaces, serialized server, and Runner), all eight browser shards, typecheck/release registry, build, canary dry run, security checks, and policy gates. There are 52 passing checks and no pending or failing checks. - Greptile completed on the exact current head with 5/5 and no actionable findings or open review threads. - Local repair verification passed 93 focused tests, including same-app reinstall after revocation and rejection of consent started before revocation, the AgentMail browser journey, and repository typecheck. Local build and Storybook build also passed. The redundant local full-suite rerun was stopped after the complete current-head CI matrix passed. - Real Slack setup and agent replies were exercised in the authorized isolated test drive during the setup iteration. - The ten additional model probes were attempted with legacy `codex_local`, `gpt-5.6-sol`: five passed, two failed, and three were partly verified. Native runtime is not qualified. See `server/src/services/connectors/slack/evals/2026-10-08-acceptance.md` for evidence and limits. - Passing model probes cover native forms, downloaded file bytes, bare mentions with thread replies, explicit posts/reactions, and saved approval denial. - The controlled uncertain-write probe found wrong delivery-check IDs. The canvas fallback attempt used an invented tool name. Search pagination/native search, a private-source denied-tool receipt, and distinct board/webhook origins remain unqualified. Reviewer path: enable Chat connectors, start Slack chat setup, select an agent, enter an app-configuration access token, and approve Slack installation. Send a message to the bot and confirm that setup advances to success. Inspect settings and allowed channels. See `doc/connections/SLACK-AUTOMATIC-SETUP.md` for deployment and recovery. ## Risks - Slack app creation has no provider idempotency guarantee. A timeout after dispatch stays uncertain until the operator checks Slack. - OAuth needs a stable public HTTPS board origin. Webhook ingress may use a separate configured HTTPS origin. Workspace policy can delay installation. - Migration 0318 can replay safely on instances that applied the earlier development migration. - OAuth installation now links the installer to the initiating Paperclip user. Identity checks and company access rules still apply. - Joined channels now enable bot responses by default. Linked-user authorization and per-action approval rules still apply. - Model behavior has the documented delivery-check and canvas fallback failures. A passing CI run does not establish that every model probe passed. - Removing the connection does not delete the customer's Slack app. No new first-party telemetry is added. ## Model Used OpenAI Codex, GPT-6 family, with reasoning, repository tools, code execution, and browser verification. The runtime does not expose a more specific authoring model ID or context-window size. The live bot probes used OpenAI `gpt-5.6-sol` through `codex_local` in legacy mode. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> |
||
|
DottaandPaperclip
|
d0f69670db |
fix(runner): recover saved execution prompts after upgrades (#15518)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Native runs save an immutable execution context for restart recovery. > - The context includes the prompt text, revision, and content hashes. > - The parser required that saved prompt to match the current release. > - A server upgrade could reject a valid saved run before provider recovery. > - This pull request validates and preserves the saved prompt snapshot. > - Routine prompt changes no longer need a catalog of past strings. ## Linked Issues or Issue Description **What happened?** A hot restart selected a dead native runner for same-run recovery. Reading its saved v5 execution input failed with `input.runtimeContext.prompt must match the fixed Paperclip prompt revision`. The new controller accepted only v6. **Expected behavior** Recovery uses the saved prompt and validates its content hashes. It preserves the same run and provider session without starting a duplicate turn. **Steps to reproduce** 1. Start a native run and save its execution input and provider checkpoint. 2. Change the fixed execution prompt in the server release. 3. Stop the runner and recover the saved run with the new controller. 4. Observe that the old parser rejects the saved prompt before provider recovery. **Paperclip version or commit** The v5-to-v6 prompt change was introduced in #15446. The defect also reproduces on current master before this fix. **Deployment mode** Source-built server with the native runner. Related work: #15446 added task-monitor guidance. The held prompt-size experiment in #15489 changes prompt wording but does not add recovery compatibility. ## What Changed - Read the prompt text and revision from the saved execution snapshot. - Treat the revision as non-empty metadata and preserve the exact saved bytes. - Validate the prompt SHA-256 and the aggregate context digest. - Keep fresh-run builders on the current prompt constants. - Test arbitrary saved prompts, malformed fields, altered text, stale hashes, and aggregate drift. - Test recovery parsing for Codex input versions v3-v5 and OpenCode, ACPX Pi, and Dot v6 inputs. - Extend the real-process restart suite with both the incident's v5 wire fixture and a prompt unknown to this release. - Run the restart recovery suite in the existing Rust-equipped PR lane, where its runner and fake-provider binaries are built. Verify complete, non-overlapping test coverage for PR, release, and local callers. - Document recovery from saved snapshots without a historical prompt catalog. ## Verification - Red: the new contract regressions fail against the catalog-based parser with the original prompt-validation error. - Green: 53 focused contract and materialization tests pass. - Red: the real-process unknown-prompt regression fails with master's original parser at the saved-input recovery read after process loss. - Green: all 15 real-process restart tests pass locally on the final branch. The saved-prompt cases keep the run and provider session, replace the PID, and record one `turn/start`. - Local repository `pnpm -r typecheck` and `pnpm build` passed after rebase on `89f09dad723766e5351953f0731b9aa5daada28d`. The 53 focused tests also passed on that head. - Red: the new test-roster checks fail against the old CI placement. - Green: all 26 test-scheduling checks pass after moving the restart suite. - CI ran all 15 restart recovery tests with no skips on final head `6719fc2bb7a31a0f72ea04c7e525a63dcc6f9105`. [Runner test job](https://github.com/paperclipai/paperclip/actions/runs/37764782334/job/113271464966). - Greptile scored 5/5 on that exact head with no actionable findings. - The complete CI matrix passed on final head `6719fc2bb7a31a0f72ea04c7e525a63dcc6f9105`: general/workspace tests, serialized server suites, both runner Vitest lanes, Rust and static checks, all browser shards, typecheck, build, and the canary dry run. [CI run](https://github.com/paperclipai/paperclip/actions/runs/37764782334). - There are no unresolved review threads or merge conflicts. - Reproduce focused tests with `pnpm --filter @paperclipai/paperclip-runner exec vitest run src/contracts/runtime-context.test.ts src/contracts/native-execution.test.ts src/drivers/runtime-context-materializer.test.ts`. - Reproduce restart tests with `pnpm --filter @paperclipai/paperclip-runner build:rust` followed by `pnpm exec vitest run server/src/services/native-runtime/native-runner-restart-recovery.integration.test.ts`. They use temporary PostgreSQL, real runner processes, and a fake Codex provider. They do not use paid inference. ## Risks - The parser now accepts internally consistent saved prompt text that is absent from the current source. Inputs must come from trusted server persistence. Content hashes verify consistency; they do not authenticate authorship. - Existing execution-schema, ownership, checkpoint, provider, permission, and session-compatibility checks still apply. - This change validates the saved base prompt. It does not make all additional code-generated instruction strings versioned. - The process-level recovery proof uses Codex. Other provider coverage verifies the shared input parser and retained provider configuration. ## Model Used OpenAI Codex, GPT-6 family, with repository inspection, code editing, and test tools. The exact serving model ID and context-window size are not exposed in this session. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>canary/v2026.1008.0-canary.7 |
||
|
DottaandPaperclip
|
bf9dbd18a8 |
fix(runner): recover incompatible Codex models before launch (#15519)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - The Runner prepares and saves each task's provider configuration before launch. > - A sandbox image can have a supported Codex CLI that is too old for the selected model. > - The current version check stops that task even when the image can run a similar model. > - This pull request selects a compatible model before it saves a fresh execution. > - The task continues with a visible warning, and recovery uses the saved effective model. ## Linked Issues or Issue Description - Refs #15053. This fixes the same CLI and model mismatch for fresh remote Runner tasks. It does not change subscription onboarding. - Refs #14721. This adds a bounded startup choice for Codex CLI compatibility. It does not add user-defined fallback chains or quota failover. - Related PR: #15399 added the model-specific CLI checks that this change preserves at launch. ## What Changed - Check the preinstalled remote Codex CLI before saving a fresh execution that uses a model with a verified CLI minimum. - Select the closest compatible older model in the same class, then the stable Runner default. Consider each candidate once. - Save the effective model before checkpoint selection. Keep the requested agent and task settings unchanged. - Add a task warning and a local `runner.model_fallback` run-log event with both models and the CLI version. - Share executable discovery with the launch verifier. Preserve explicit artifact and install settings, saved executions, and existing artifact checks. - Add regression tests and document the selection and warning behavior. ## Verification - Red: the three new provider-configuration regression cases failed before the fix. They kept the incompatible requested model. - Red: both rejected-probe cases failed before the review fix. They now defer to launch verification. - Green: targeted provider configuration, remote preflight, task warning, and launch-verifier tests pass (680 tests). - `pnpm -r typecheck` passes. - [Full CI](https://github.com/paperclipai/paperclip/actions/runs/37713759744) passes on `eb273fc32661526d90a864b42c73a7da61dd62da`: 47 successful jobs, including all general and serialized Vitest groups, browser shards, Runner verification, typecheck, build, and the canary dry run. - Local extended verification: three serialized shards pass (115 suites). HTTP route tests hit intermittent 15-second timeouts. The authorization suite passes on rerun (130 tests). The document suite passes on both master and the PR head in the same isolated setup (6 tests). The local general run was stopped after the equivalent full CI matrix passed. - `pnpm build` passes. - Server typecheck and build pass again after the probe-error review fix. - `git diff --check` and `pnpm check:module-boundaries` pass. - Example: request `gpt-6.1-sol` on Codex 0.158.0 to use `gpt-6-sol`; on 0.156.0, use `gpt-5.6-sol`. The warning names the requested model, effective model, and CLI version. - This PR has not been deployed to staging. ## Risks - A fallback can have different capabilities. The task warning makes the substitution visible. Later runs can use the requested model after the image CLI is updated. - Preparation adds two remote commands for models with a verified CLI minimum. Failed or invalid version probes retain the existing launch checks. - This only handles known CLI and model mismatches before a fresh provider launch. Authentication, capacity, and artifact failures keep their existing behavior. - No database migration is required. ## Model Used OpenAI GPT-6 through Codex, with tool use and code execution. This session does not expose the exact deployment model ID, context window size, or reasoning setting. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either (a) linked existing issues with `Fixes: #` / `Closes #` / `Refs #` OR (b) described the issue in-PR following the relevant issue template - [x] I have not referenced internal/instance-local Paperclip issues or links (only public GitHub `#NNN` / `github.com/paperclipai/paperclip` URLs) - [x] My branch name describes the change (e.g. `docs/...`, `fix/...`) and contains no internal Paperclip ticket id or instance-derived details - [x] I have run tests locally and they pass - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>canary/v2026.1008.0-canary.6 |
||
|
DottaandPaperclip
|
b5342febe5 |
fix(runner): require current-turn completion after connection continuations (#15514)
## Thinking Path > - Paperclip manages agents, tasks, permissions, and execution budgets. > - Native tasks can resume after a connection decision in the same provider conversation. > - Each turn still needs an accepted completion report. > - Compact continuation messages did not explain that reports from earlier turns cannot finish the new turn. > - The connection evaluator could also grade before the final reply was stored or reject valid unavailable-access wording. > - This PR clarifies the current-turn report requirement and fixes those observation boundaries. > - The original connection instructions and strict native completion gate stay in place. ## Linked Issues or Issue Description Refs #15489. The reduction remains draft while this separate repair is qualified. Refs #15471 for the earlier connection continuation work. ## What Changed - Add a current-turn completion reminder to compact continuation inputs. - Keep final prose insufficient for completion. Preserve permissions and retry policy. - Wait for the final successful task run's saved, attributed decline reply within the existing deadline. - Use one bounded explanation matcher for both decline checks. - Wait for a recorded tool-action rejection to dispatch its bound continuation, with strict company, task, agent and source-run checks. - Retain the exact grading input before later API refreshes. - Add failure and delay regressions and update the Runner and evaluator docs. ## Verification The fresh comparison has **15/15 original passes on each variant**: 15 unchanged pass pairs, zero new failures, and no pending pair. There are 30 case attempts and **65 actual agent runs** (baseline 33; candidate 32). All runs succeeded. All 30 cleanups passed. No model attempt was retried. | Profile | Baseline | Candidate | | --- | --- | --- | | Native Codex `gpt-5.6-sol` | 5/5 | 5/5 | | ACPX Claude `claude-sonnet-5` | 5/5 | 5/5 | | OpenCode `openrouter/deepseek/deepseek-v4-flash-0731` | 5/5 | 5/5 | Each profile covers service approval, service decline, connection decline, provider decline, and selection of the second provider. The saved replies, approved briefings, decisions, fixture observations, final task states, and native completion records were inspected. All 18 saved decline-grade snapshots match their original captured inputs and checks. Result, API snapshot, and final ledger run sets agree. - [Candidate campaign](https://github.com/paperclipai/paperclip/actions/runs/37711658378) · [public candidate report](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-37711658378-1/index.html) - [Baseline campaign](https://github.com/paperclipai/paperclip/actions/runs/37711675579) · [public baseline report](https://d1p6rlowie26tp.cloudfront.net/runner-e2e/campaigns/gha-37711675579-1/index.html) - Candidate source: `79905343bba280d462765faad19a26e7f179259e`. Baseline source: `7c5e120158f1385a1fc5f20be41f66c58a605534`. Both use master context `fc6304dfe5f2e446e09bd052a7b45f51e930f250`. The trusted workflow source is separately frozen at `dd777f4b7343305c4e6f44c422f44a1d78e12e4f`. - Both variants use the same evaluator, fixtures, models, permissions, 720-second cell deadline, and 1,000-cent company and agent hard stops. The only production difference is the compact continuation reminder. - Suite hash: `aee30b74b4d38ada08777798db0932fbc64e368bb427d29caa8cfd86c7f59747`. Definition hash: `ea9e17f54fe0af1acbb2337ddfab8a7e92db8488bd0f59510a63095ca0229b60`. - Provider-free transport capture: startup/resume input stays at 55,726 bytes. Compact continuation input grows from 53,334 to 53,535 bytes. The 42-tool catalog stays unchanged. These are Paperclip input bytes, not complete vendor prompt tokens. - Focused evaluator tests: 31 pass. Native contract, transport delivery, and session tests: 194 pass. Evaluator support: 1,819 TypeScript tests and 128 Node tests pass, with one intentional skip. - Full build, workspace typecheck, and evaluator typecheck pass. Current-head CI passes all required gates. The current rollup has 51 successful check runs, two intentional Storybook skips, and a successful Snyk status. Review is 5/5 with zero unresolved threads. - CI attempt 1 had one initial runtime-fixture health timeout. The exact test and its full 164-test file pass locally. One CI shard retry passed. The original CI failure, its dependent verify failure, and the retry remain visible in [CI history](https://github.com/paperclipai/paperclip/actions/runs/37711235060). - The broad local `pnpm test:run` attempt was interrupted after about 49 minutes (exit 130). It recorded one failure in the unchanged Zep memory-connector disabled-setting test. That test and the full 388-test tool-access file pass in separate local checks; current-head CI also passes. The local cause is not established, and this broad local attempt is **not** claimed as passing. Three earlier local failures also pass in their isolated checks; their original logs remain retained. - The first two setup admissions were cancelled before provider jobs to include the review correction. They made no provider calls. The completed campaigns above are the first and only model attempts for these corrected variants. Cost evidence stays separate from behavior. Original result summaries report only OpenCode amounts: baseline $0.039192096 and candidate $0.063221620. Final run ledgers also retain estimates for Claude (baseline $1.232439000; candidate $1.333611200) and Codex (baseline $2.340324400; candidate $2.129335600). These estimates do not replace the original summaries. Local and GitHub runtime are unmetered here. Invoices are unknown. This is not a cheaper or faster claim. ## Risks - A single matched trial cannot prove general equivalence or causation. The reminder is an instruction change, not a new completion enforcement rule. - Candidate OpenCode service-decline finished within one continuous run; its baseline used two. That pair passed the task outcome, but it does not qualify the reminder on a resumed decline turn. No extra paid run was used to replace it. - The text matcher is bounded evidence of an explanation. It does not prove reasoning or consumption of feedback. Bounded stdout excerpts do not prove that every extra attempted tool call is absent. - Missing saved replies or continuations still fail at the original deadline. Failed native completion remains a failure even when final prose is correct. - The unresolved local-suite discrepancy above remains a validation limit. Full remote CI and both focused local reproductions pass. - The original connection instructions stay in place. These results do not qualify the reduction in #15489. Its original 11/15 versus 12/15 grades and two new failing pairs remain unchanged. ## Model Used OpenAI Codex, based on GPT-6. The exact deployment ID and context window are not exposed in this session. Capabilities used: reasoning, repository editing, code execution, test inspection and eval analysis. ## 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 linked existing issues or described the issue in-PR - [x] I have not referenced internal/instance-local Paperclip issues or links - [x] My branch name describes the change and contains no internal Paperclip ticket id or instance-derived details - [x] I have run the focused local tests listed above and they pass; the interrupted broad local run is disclosed above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing>canary/v2026.1008.0-canary.5 |
||
|
Devin FoleyandPaperclip
|
5717523b9e |
fix: retain the original repository for reused task workspaces (#15528)
Apply the reviewed change for fix: retain the original repository for reused task workspaces. Validation: required local typecheck and tests, passing CI, Greptile 5/5, and independent review of the exact source head. Co-Authored-By: Paperclip <noreply@paperclip.ing>canary/v2026.1008.0-canary.4 nightly/v2026.1008.0-nightly.0 |