mirror of
https://github.com/paperclipai/paperclip.git
synced 2026-11-03 01:27:09 +01:00
4e97daafa06a54f02027335437c4f48e50b84236
14
Commits
| Author | SHA1 | Message | Date | |
|---|---|---|---|---|
 DottaandPaperclip
|
f1911a4c53 |
perf(skills): move repository packages out of Postgres (#18082)
<!-- entire-trail-link-start --> https://entire.com/gh/paperclipai/-/trails/205/changes/paperclip/205 <!-- entire-trail-link-end --> ## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents receive selected company skills before each turn. > - Imported skills can need files from the whole source repository. > - The current path stores those files in Postgres and reads large JSON payloads during preparation. > - This pull request stores verified packages in the configured file or object store and keeps preparation queries small. > - It also converts old rows automatically and restores missing files from the exact saved commit. > - The result reduces database traffic while preserving skill versions and repository dependencies. ## Linked Issues or Issue Description **What existing behavior does this improve?** Company skill import, version storage, and preparation before agent turns. **Current behavior** Repository snapshot bodies live in JSONB. Skill preparation can read those bodies again even when the files already exist on disk. Historical packages and native assets also need a bounded cleanup path. **Proposed behavior** Store package archives and manifests through the existing storage service. Keep version identities, receipts, hashes, and recovery metadata in Postgres. Convert legacy rows after the stored package passes read-back verification. A missing package can use legacy bytes or its exact Git commit under current repository authorization. No separate data-migration command is required. **Reason and benefit** The warm path sends no repository bodies through SQL and reads no object bytes. Cleanup runs through the existing deadline coordinator. Current versions, active runs, explicit pins, and authored content retain their protection. **Breaking changes** The normal application schema upgrade is required. Single-version and skill-file responses keep their existing shape. Version lists, current-version summaries, and test-run version summaries now contain file names and kinds without file bodies. Diff and restore fetch selected versions through the existing single-version endpoint. Older server builds must not write after conversion unless they support the new package format. The storage override and rollback limits are documented in `doc/skill-package-storage.md`. **Additional context** Related work: #15731 preserves repository dependencies; #17923 changes version-cache scheduling; #18080 changes delivery to remote Codex homes. This change owns package persistence, acquisition, and expiry. It does not replace those delivery or scheduler changes. ## What Changed - Add company-scoped archive receipts, exact-commit recovery recipes, migration deadlines, and a durable object-deletion queue. - Add bounded archive upload, read-back verification, extraction, concurrent request coalescing, and verified local caches. - Convert old snapshots and versions automatically. Keep legacy bytes when verification fails. - Recover missing files through current source authorization and the original commit. Validate repository identity, file hashes, modes, and the aggregate hash. - Expire unused recoverable Git packages and native assets after 30 days. Preserve active pins and the durable copy of authored or unproven content. - Keep warm preparation queries metadata-only. Bound file digest caches and avoid repeated body hashing when file identities are unchanged. - Extend the existing storage contract with cancellation. Register package work with the existing coordinator. Optional conversion retries retain the original bytes and may wait for ingress without blocking sleep. - Add regression tests, operating guidance, a deep trace, benchmark receipts, and an optional maintenance command. ## Verification ### Final master integration and review repair — October 29, 2026 Current head: `a6065424249e763dfeac37161f1b52a9f25bd9b7`. Commit `f8bbdd3dbd` resolves the conflicts with master `76a796e964`; the final two commits repair the Apex deletion findings below. - Keep both package-maintenance and version-retention notifications in their write transactions. - Keep the shared run-pin predicate and master's controller-lease deadline metadata. Package expiry reads the updated protected-version result. - Preserve every master migration entry. Generate migration **0415 (`0415_zippy_zemo.sql`)** after master's 0414. Its skill SQL is byte-for-byte unchanged from the previously tested idempotent migration. Retain only the five newest schema snapshots. Migration numbering and safety checks pass. - Repair failed file/folder deletion: preserve a complete temporary subtree until transaction commit, restore on upload or SQL failure, and restore callback failures before releasing the mutation lock. Both PostgreSQL regressions reproduced missing files before the fix and now pass. They cover binary data, executable permissions, empty folders, unlisted files, unchanged database/version rows, successful retry after upload-claim expiry, and historical reads. - Reconcile lost commit replies under the same advisory and row locks. Check the saved version and source identity before restoring. Keep committed deletions intact and preserve backups when the database outcome is unknown or a newer edit exists. Four real-PostgreSQL outcome tests pass; the prior code failed three of them. - Full workspace typecheck and production build pass on the final commit. All **204** focused PostgreSQL package/cache/retention/authoring and filesystem-boundary tests pass, with none skipped. - Apex review **18 passes 5/5**, with zero new comments on the exact final head. All **34** review conversations are resolved. - [Final-head CI](https://github.com/paperclipai/paperclip/actions/runs/40914116508) completes successfully on attempt 2. The head has **55 successful checks**, two neutral reports, two skipped checks, and a successful Snyk status. No checks are pending or failing. All eight browser shards pass. - Runner Vitest 2/2 timed out once in the unchanged ACPX controller-replacement fixture. The same job passed on the preceding commit. One unchanged CI retry passes all five ACPX recovery cases, **2,253** Runner package cases and **126** native server integration cases; their standard one and five skips remain. The initial timeout's root cause is not established. Local diagnosis also hit the fixture's eight-second session.open timeout with an authenticated Runner; no local Runner pass is claimed. - Both full local `pnpm test:run` attempts were stopped when Apex identified deletion defects and source repair became necessary; neither is claimed as a passing full local run. The final **204** focused tests, full workspace typecheck, and production build pass. Full-suite acceptance uses the final-head CI jobs. The superseded merge-head CI included runners terminated by shutdown signals; each source repair started a new full run. - Auto-merge is enabled for squash merge. The final head is conflict-free and all checks pass. GitHub's active master ruleset still requires a code-owner approval for the dependency files. The existing review requests are with `devinfoley`, `nickyleach`, and `forgottendev`; no policy bypass was used. - The earlier hosted stress and full native receipts below remain tied to `a2e597d800`; they are not relabeled as a deployment of this final integration. The earlier local PostgreSQL startup failure remains recorded below; PostgreSQL ran successfully for this integration verification. ### Full native skill execution — October 29, 2026 The full native-agent acceptance gap is closed on candidate `a2e597d800eb989d565780e7aa947d5368676444`. Two qualifying native turns used OpenAI API / `gpt-5.4-mini` in real Daytona sandboxes, on the same disposable hosted TEST tenant as the stress test. A follow-up on the same task consumed a newly saved skill version. The task text asked for the assigned skill workflow. It did not contain the proof nonce or expected receipt. The assigned authored skill contained a Python verifier and a random package-local nonce. The agent read the actual installed skill files and executed that verifier against the imported, automatically converted 1,360-file repository package. It checked the repository file count and reference/binary bytes, and directly executed the repository's executable probe. Independent verification matched the emitted JSON to the nonce and expected byte hashes. The agent persisted the same receipt as its task response; native finalization committed, workspace finalization succeeded, and task state became Done. | Qualifying turn | End-to-end run time | Recorded skills.prepare | Repository files | Binary hex | Shell receipt SHA-256 | |---|---:|---:|---:|---|---| | Initial task | 47.881 s | 1822 ms | 1360 | `00ff07800a` | `a1deb30dc095f2306c7aa9cc5ec396da12c68961cf5b809f4d7c20f7fc68493b` | | Corrected follow-up, new nonce/version | 37.974 s | 377 ms | 1360 | `00ff07800a` | `dbcdc33fa1376b4edc136feae1cdf770560a73292467e93619efdda20e4282a7` | Both successful shell receipts reported `probe: STRESS18082` and `assetSha256: e8ea7a2f15f7bb3b332a46adbef848465884c8996595aa234748befd544d1f3a`. Their proof nonces, selected version IDs and receipt hashes differed, and both ran on the same task. These are observed timings, not latency SLOs; they include real provider work. The service stress-test timings remain a separate measurement boundary. Earlier failures remain in the evidence: - The original five-minute Daytona allocation timeout never reached model execution. After normal owned-stack restore, its durable cleanup record was `expired` / `success`. The cause of that original allocation timeout remains unknown; a later success establishes recovery only. - Initial continuation setup used unsupported generic agent-update fields and returned HTTP 409 before dispatch. Setup was corrected to use the supported skill-assignment and versioned instruction commands. - The intervening completed native turn read repository bytes and ran the probe, but omitted the verifier JSON under the older test agent's overly-specific instruction file. Its strict receipt validator failed; it is not counted as a qualifying turn. Correcting the test agent's instruction file and sending an explicit task-scoped wake produced the qualifying follow-up above. - The pinned application's immutable rate card omits `gpt-5.4-mini`, so completed usage triggered the normal block-on-unpriced budget policy. Supported append-only accounting corrections estimated the three completed runs using the [official OpenAI model rate card](https://developers.openai.com/api/docs/models/gpt-5.4-mini): $0.75 input, $0.075 cached input, $4.50 output per million tokens, assuming standard processing. Estimates totaled 12.482655 cents. Original usage and missing-price records are preserved; the $1 hard stop and unpriced-usage guard stayed active. - Model shell attempts included a missing `xxd` command and incorrect guessed file paths. Those failed calls and subsequent successful actual verifier calls remain visible in complete event streams. No zero-tool-error claim is made. No product source changed during this continuation. The previously measured contention, cold-start and memory limits remain. The later merge resolution is covered by the separate integration checks above; these hosted receipts qualify their recorded candidate. ### Hosted stress test — October 29, 2026 **Hosted acceptance:** full native-agent execution and a changed-skill follow-up pass on the recorded hosted candidate. The final integration section above records the subsequent conflict resolution and its separate checks. Tested baseline `929e840595c8933ca485ac7a2e99413e1e11ca64` and this exact candidate sequentially on one disposable TEST tenant, with external Neon PostgreSQL and actual S3. The fixture contained **1,360 files / 22.4 MB / ten skills sharing one repository**. Measurements cover deployed skill-preparation and native-context services in an adjacent diagnostic process, not a complete agent turn. The container had **24 vCPU / 24 GB**. | Warm workload | Median before → after | p95 before → after | DB returned value bytes/preparation before → after | |---|---:|---:|---:| | One skill, serial | 1,060 → 125 ms | 1,303 → 155 ms | 22,610,389 → 1,771 | | One skill, four concurrent | 4,415 → 297 ms | 5,143 → 328 ms | 22,610,389 → 1,771 | | Ten skills, serial | 5,723 → 535 ms | 6,057 → 600 ms | 224,739,030 → 10,671 | | Ten skills, four concurrent | 24,410 → 1,986 ms | 24,833 → 2,221 ms | 224,739,030 → 10,671 | - Candidate warm preparation performed **zero S3 GETs**. Ordinary-pool queries fell from 16 to 9 for one skill and 61 to 9 for ten, excluding dedicated lock connections. DB numbers measure returned values before Drizzle decoding, not wire bytes. Baseline/candidate sample counts were 30/100, 32/100, 5/50 and 8/48 respectively; the small baseline ten-skill samples are comparative evidence, not SLO estimates. - **Application-UID soak:** 800 ten-skill preparations over 10.3 minutes, four workers paced by 2.5 seconds; **zero failures**, median **559 ms**, p95 **775 ms**, p99 **910 ms**, zero warm S3 reads. All 726 accompanying health/company-list requests returned HTTP 200. Their end-to-end p95 was 569/502 ms, including network latency. The probe overlaps most of the soak. Soak DB-response capture was limited to its first 60 seconds. - **Important remaining costs:** an unpaced 16-way ten-skill burst had p95 **8.32 seconds** because company-level preparation serializes. Cold preparation took approximately **1.1–2.5 seconds** for one skill and **7.4–10.0 seconds** for ten. Ten-skill diagnostic RSS remained about **1.17 GiB** (baseline 1.18 GiB), and the soak peaked at **1.35 GiB**. A concurrent whole-container sample was 3.45 GB. This does not establish suitability for a small-memory host or eliminate per-entrypoint native asset duplication. - **Upgrade without a manual data migration passed:** legacy snapshot JSON shrank from 22,465,691 bytes to `[]`, with a stored package and zero pending conversions. - **Recovery passed:** sixteen concurrent cold requests coalesced into two S3 GETs; full bytes/binary/executable modes matched; same-size corruption repaired; an injected storage failure preserved warm access and failed cold access safely. A disabled public-Git source denied recovery, and after the actual five-minute backoff its exact SHA reconstructed all 229 files. Cross-company descriptor access was denied. Private-Git credential revocation was not tested. - **Retention passed:** reclaimed 27,754,496 allocated cache bytes, deleted an unreferenced package row and both objects, preserved aged authored history, and expired/reacquired reconstructible Git content. A version pinned by the failed native run was correctly retained. Only owned fixture timestamps were aged. - **Browser UI passed:** authored skill creation, saved edit surviving reload, version history, and restoration of earlier content. - **Initial native attempt (superseded by the qualifying turns above):** one real run failed at Daytona allocation after five minutes (`setup_failed`), before any model execution. Provider cleanup also reported failure. That attempt does not count as a passing native turn. Its receipt and all evidence remain preserved; its cleanup later recorded success after normal restoration. - Initial A/B commands ran as SSH root. Diagnostic-owned lock/cache permissions caused one setup HTTP 500 and the first UID-1000 soak to fail; exact fixture paths were repaired. The final soak, retention and recovery ran as application UID 1000. A fresh UID-1000 one-skill batch also passed (30 samples, median 116 ms, p95 321 ms, 1,771 DB value bytes/preparation). Earlier harness-export, SSH, npm provenance, backoff and UI-wait failures remain recorded alongside their successful retries. - Private evidence includes raw timings, assertions, HTTP probes, screenshots and failure/cleanup receipts. Initial failed-campaign cleanup suspended and retained the owned TEST tenant and released the shared QA account lease at **15:22:41 UTC**. The tenant was subsequently resumed under its existing ownership for the native execution qualification above. Final cleanup completed at **15:53:09 UTC**: the account lease is released and the tenant is suspended/retained under the original October 30, 12:24:52 UTC retention deadline. Its separate stack lease remains held; no active or pending-cleanup environment lease remained. The original retention window is deliberately preserved even after successful continuation. - Earlier hosted-tested head: `a2e597d800eb989d565780e7aa947d5368676444`, integrated with master `929e840595`. The earlier merge conflicts were resolved at this head. [Current-head CI](https://github.com/paperclipai/paperclip/actions/runs/40884087629) passed; all checks are complete with 55 successes, two neutral reports, two skips, and no failures. [Apex review 15](https://github.com/paperclipai/paperclip/pull/18082#issuecomment-6480257258) passed 5/5 with zero findings on this exact head. All review conversations are resolved. - Ten real-filesystem boundary cases and server typecheck pass locally for the security fix. Coverage includes cache roots, descendants, individual files, symlink aliases, recursive ancestors, a future concurrent cache fill, and allowed sibling directories. [Current-head private-project integration tests](https://github.com/paperclipai/paperclip/actions/runs/40884087629/job/123483306405) passed all 29 cases, including authorized and unauthorized imports of a cached private package. Full typecheck, build, server suites, all eight browser shards, and Runner checks passed in current-head CI. Local database tests could not run because PostgreSQL could not initialize; this limitation is recorded below. - The earlier generated migration preserved the newest five snapshots. The final integration renumbers its unchanged SQL to 0415. Migration safety/numbering checks pass. Its SQL is byte-for-byte identical to the previously verified idempotent skill migration; repeated application on this latest schema was blocked by the same local database startup failure. - The broader verification before the final migration-only rebase passed 1,025 package, lifecycle, remote-launch, native-session, and asset-retention tests, 52 focused UI tests, full workspace typecheck/build, token gates, and map checks. The linked complete CI run passed before the latest focused fixes. - A disposable MinIO instance passed multipart upload, verified read-back, and cold restoration of an 18 MB archive. - Twenty warm samples with a synthetic 1,349-file, 22.8 MB repository retained five metadata queries and zero object reads. Median preparation was 256 ms for one selected skill and 691 ms for ten, with 5 ms emulated network delay each way. These are local measurements, not production latency guarantees. - No production instance was changed. Earlier verification failures and their disposition are retained below. <details> <summary>Earlier failures, review findings, fixes, and rerun evidence</summary> - The first CI pass found a stale inventory-span expectation and a missing repository-read span. Both are fixed; all three observability tests pass. One browser test copied page text into its YAML parser after reload; its first unchanged-head retry passed, but the race recurred after rebasing. The test now waits for the renamed URL and completed reload before copying the editor, retaining its original assertions. The earlier full local suite did not complete cleanly: it reported setup timeouts and assertions, while isolated relevant reruns passed. The complete current-head CI run closes that broad verification gap. No production instance was changed. - The first Apex review scored 2/5 and identified six issues. The fixes protect authored history even when it shares an in-flight Git upload, recheck upload expiry during orphan deletion, bound failed-upload cleanup, retain per-version kind/encoding metadata, add sanitized failure diagnostics, and correct the migration guide. The final focused run passes all 140 tests; the preceding eight-file run passed all 149 tests, including local storage and S3 diagnostics. Server typecheck and migration safety/numbering checks pass after the fixes. The second Apex review scored 1/5 and found unbounded maintenance storage calls and mismatched package/Git batch limits. Those are fixed with a 30-second cancellable cleanup deadline, shared 100 MiB/10,000-file scanner and storage limits, and bounded Git read batches. The proposed manual lockfile edit was rejected with evidence of the enforced CI-owned lockfile policy. All 19 storage/recovery unit tests and server typechecking pass. The new database-backed cleanup/conversion tests were locally skipped because the host exhausted its PostgreSQL semaphore capacity (confirmed directly by initdb). That gap is closed by current-head CI: [package-store tests](https://github.com/paperclipai/paperclip/actions/runs/40771320775/job/123109300889) pass all 18 cases and [package-worker tests](https://github.com/paperclipai/paperclip/actions/runs/40771320775/job/123109301157) pass all 7, with no skips in either suite. The corrected YAML browser shard passes. A separate planning-fixture test failed its cleanup DELETE with an activity-log foreign key violation after its behavior assertions passed; its one unchanged-head retry passed. The third Apex review scored 1/5 and identified a one-slot pool deadlock, projected path limits, cleanup holding the preparation lock during archive reads, and a publication/expiry race. These are fixed with transaction-local reads, prefix-aware bounds, isolated verification, and shared package row locks with a publication grace. The separate Paperclip review also found stale historical inspection, eager history downloads, and conversion retries blocking sleep. Those are fixed. Local PostgreSQL capacity recovered: the new database regressions ran, including the one-slot fork, paused storage during a warm turn, and a locked publication during expiry. The expanded run passed 144/145 tests; its remaining assertion still expected file bodies from the version list and was updated to fetch the selected version. The corrected company-skill suite passes all 81 tests, and the final package/projection run passes all 27. All 52 focused UI tests pass after updating their query-provider fixtures; the first UI run exposed a fixture remount, and the follow-up exposed an editing syntax error, both corrected. Full workspace typecheck and build, token gates, and map checks pass. The sandbox initially blocked the build tool’s local IPC socket; the authorized build rerun passed. Master then claimed migration 0401. The branch is rebased on |
||
|
DottaandPaperclip
|
30b753f614 |
perf(budgets): scope accounting and deliver enforcement asynchronously (#17046)
## Thinking Path > - Paperclip manages AI agents and their paid work. > - Budgets must prevent new work and stop reported overruns. > - Cost settlement also ran serial lifecycle reconciliation for every eligible company agent. > - That made request latency grow with unrelated idle agents and delayed results on stop delivery. > - This change separates atomic admission and accounting from durable asynchronous enforcement. > - Scoped live usage detects overruns without writing provisional ledger charges. > - The result is constant query scope for ordinary settlement and recoverable stop delivery. ## Linked Issues or Issue Description **What happened?** Settlement and budget checks entered company-wide lifecycle reconciliation. A company with 1,000 unrelated idle agents performed 1,001 accounting transactions for one ordinary cost evaluation. User requests could also wait for provider termination. Running usage checkpoints did not enforce reported monetary exposure. **Expected behavior** Keep the final reservation gate authoritative. Commit charges, scope blocks, and enforcement intent before returning. Deliver stop requests after commit. Read only applicable policies and execution scopes for ordinary charges. Stop long-running work when supported reported or priced usage establishes an overrun. **Steps to reproduce** 1. Create a disposable company with one company budget and idle agents. 2. Evaluate a charge below the budget with 0, 17, and 1,000 unrelated idle agents. 3. Compare query counts, company accounting transactions, and duration. 4. Use the controlled partial-usage and delayed-stop fixtures to check enforcement and request responsiveness. **Paperclip version or commit** The measured baseline is `6c9b87d0922f1ebfc43e8b0c36889ac5ba7b9837` from `master`. Related public work: Refs #16976, #2006, #3539, #13763, #14296, #16853, #13075. These cover the affected service charge, cancellation, and accounting paths. This change removes their shared reconciliation overhead. ## What Changed - Batch applicable policy exposure under the existing company accounting lock. Preserve exact fractional money and zero-reservation dispatch fences. - Commit a durable final budget dispatch claim before provider I/O. Retain a separate final issue/run ownership transaction through synchronous handoff, without holding the company accounting lock or awaiting provider work. Preserve native recovery lease checks and duplicate handoff fencing. Browser handoff rechecks current policy; it reuses a consumed rate slot only for the identical policy generation. - Separate ledger spend, live observed spend, and effective held capacity. Use the maximum of reservation and live observation for each unfinished operation. - Deliver budget cancellation through the existing coordinator. Revalidate generations and persist queued/running stop fences before external termination. Bound stop concurrency to four. - Apply agent holds only to the affected agent. Project budgets leave other projects running. Project real company transitions in bounded lifecycle batches. - Replace periodic broad reconciliation with durable pending work, startup repair, receipt replay, and concrete retry/window deadlines. Retain drain and idle-safety checks. - Enforce accepted cumulative usage checkpoints without provisional charges. Final accounting atomically replaces live exposure and settles reservations. - Add operator-gated budget spans, regression coverage, map updates, and a real-PostgreSQL comparison harness. No schema or first-party telemetry change. ## Verification - Full workspace `pnpm -r typecheck` passes on the final ownership fix, rebased onto `bbcf5dfe98`. Production `pnpm build` passes on final head `4c5f4de8b1`. Map, module-boundary, token, and Storybook catalog checks pass. - Focused real-PostgreSQL coverage includes the 292-case rebased accounting/dispatch run, 63 browser cases, 52 run-control/cancellation cases, 42 policy-generation cases, 77 lifecycle/commentary cases, and the exact Muse confirmation regression. Native executor unit coverage passes all 768 cases. Final ownership coverage passes: 74 adapter/use-case cases, nine public regression cases, and the additional deleted-project race. These exercise competing ownership, budget changes and charges between phases, zero/nonzero duplicate claims, rollback, recovery contention, read-only capacity rechecks, and controlled HTTP startup after commit. Another 74 pure/startup/tracing cases pass. The complete CI matrix and both 5/5 reviews passed on `1ac9b58`; its external security scan identified a stale-owner handoff race. The final fix restores synchronous issue/run ownership fencing after the accounting transaction commits. Final head `8e84f55ce8` has fresh Greptile 5/5 with zero new comments, Paperclip review 5/5, and a passing security scan. The complete latest-head [CI matrix](https://github.com/paperclipai/paperclip/actions/runs/39535189808) passes, including budget, browser, Runner, build, and canary jobs. The unchanged Runner close fixture passed on its sole same-head job retry. - Broad local `pnpm test:run` invocations were stopped before required source updates (rebase and review fixes). They are not claimed as full passes. Final full-suite coverage comes from the complete latest-head CI matrix; local focused coverage and the incomplete broad invocations are reported separately. - Reproduce performance with `pnpm --filter @paperclipai/server exec tsx ../scripts/benchmark-budget-enforcement.ts --baseline-ref=6c9b87d0922f1ebfc43e8b0c36889ac5ba7b9837`; repeat with `--reverse`. The harness uses disposable real PostgreSQL, 20 warm samples per variant/count, and records raw timings, SQL/lock counts, agent reads, unchanged agent state, and query plans. Final production-source measurement revision: `eff713a4f72c7f4b7a4002a701cb481d4bfccd95`; the extended harness is committed in `4c5f4de8b1`. | Unrelated idle agents | Baseline median / p95 | New median / p95 | Baseline SQL / locks | New SQL / locks | New agent reads | | --- | ---: | ---: | ---: | ---: | ---: | | 0 | 1.80 / 2.33 ms | 2.99 / 3.88 ms | 5 / 1 | 7 / 1 | 0 | | 17 | 15.53 / 19.55 ms | 2.24 / 2.60 ms | 90 / 18 | 7 / 1 | 0 | | 1,000 | 978.35 / 1,790.43 ms | 2.16 / 2.52 ms | 5,005 / 1,001 | 7 / 1 | 0 | - These measurements use local PostgreSQL 18.1 and Node 26.4.0 arm64. They isolate the cost-evaluation core and complete admission path, without hosted-region or provider latency. Prior-revision reversed-order controls also preserved the constant query counts; the final measurements above use the final production source. - Complete reserve → final ownership gate → immediate offline provider result is measured separately, with one responsible agent and one active operation at each unrelated-agent count. Admission p95 baseline/new: 0 agents **11.54 / 29.77 ms**; 17 agents **9.31 / 27.62 ms**; 1,000 agents **24.51 / 74.05 ms**. Baseline admission was already scoped; the stronger final fence adds bounded overhead. New counts stay constant at 53 Drizzle statements, three transactions, two company-accounting locks, and five agent-reading statements bound to the responsible agent. Zero unrelated-agent reads, exactly one provider callback per sample, and no leftover runs or holds. All admission scenarios stay below 250 ms. No external provider calls are included. - The partial-usage/delayed-termination fixture passed 20/20: receipt persistence p95 13 ms, receipt return to observed durable Stop fence p95 28 ms, complete receipt-start-to-fence p95 50 ms, unrelated database request p95 1 ms while termination was blocked. After releasing the controlled termination fixture, acknowledgement p95 was 14 ms. These are separately measured phases; fixture acknowledgement is not live-provider termination latency. - Models and provider deadlines are unchanged. No deployment, live inference, or staging acceptance is claimed. A quiet 30-minute follow-up waits for the merged revision on staging, then repeats classifier and controlled budget-stop acceptance in staging, recording hosted bookkeeping, queue delay, and actual provider acknowledgement separately. - Development failures retained: finalizer notification regressions (fixed; 27/27 affected cases passed), invalid browser admission fixture assumptions (fixed; browser coverage now passes 63/63), and first-head lifecycle/dispatch CI regressions (reproduced and repaired). A prior-head Cases E2E test hit its unchanged announcement-dismiss handler race while opening a filter; the same journey passed on `1ac9b58`. The external security review then found a stale-owner race between accounting commit and provider entry; final verification covers a competing run in that gap while retaining provider I/O outside the accounting lock. Early PostgreSQL startup attempts hit local semaphore pressure; serial reruns succeeded. The final-head Runner lane had one failure among 1,823 tests: the unchanged `timeout-attachment` fixture failed at its first process close after 6.28 seconds. The sole same-head job retry passed all 1,819 active tests (four skipped), including that fixture in 1.85 seconds. Three isolated local probes could not run correctly because the Rust fake-provider fixture binary was absent; cleanup masked the original startup failure. CI builds that fixture through its normal lane. These probes are inconclusive and are not counted as passes. ## Risks - This changes accounting, lifecycle, and cancellation ordering. Durable generations, source/sequence checks, ownership leases, and company locking remain required for correctness. - Provider stop acknowledgement can be slow or fail. Stop intent remains pending until existing termination proof succeeds; cancelled row status alone does not acknowledge delivery. - Mid-run limits depend on provider usage reports or supported price calculations. Providers that report only at completion cannot provide exact mid-call detection. - Signals are process-local wake hints for the supported single-active-writer model. Durable state and startup recovery remain authoritative. - There are no migrations, new queue tables, or cached balances. Repeat the classifier and controlled budget-stop acceptance after normal rollout. ## Model Used OpenAI Codex, based on GPT-6, with reasoning, tool use, code execution, and parallel coding agents. The host does not expose the exact backend model revision or context-window size, so those values are not asserted. ## 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
|
cb2d6e80d3 |
perf: stop polling idle Muse maintenance (#17000)
## Thinking Path > - Paperclip manages AI agents and their work. > - Muse connects an external worker to a Paperclip agent. > - Its maintenance loop scans the database every five seconds, even without bindings. > - These empty scans prevent an otherwise idle database from sleeping. > - The shared delivery coordinator can react to committed writes and saved deadlines. > - This change uses that coordinator and retains recovery for unfinished assignments. > - Idle Muse maintenance can now stop querying the database. ## Linked Issues or Issue Description **What existing behavior does this improve?** Muse qualification, OAuth authority, and terminal assignment maintenance. **Current behavior** A claimed instance runs two empty scans every five seconds even when Muse has no bindings. This produces about 34,560 unnecessary queries per day. Warm standby suppression from #16974 does not cover claimed instances. **Proposed behavior** Recover once at startup. Wake after relevant transactions settle. Schedule qualification expiry at its saved deadline. Retain five-second recovery only for failures and unfinished assignments. **Reason and benefit** Remove recurring database traffic from idle Muse maintenance while preserving authority checks and recovery. Reuse the database work signal foundation from #15625 without adding tables. ## What Changed - Register Muse maintenance with the shared delivery coordinator and its idle admission rules. - Signal pairing, OAuth attachment and revocation, qualification changes, and new assignments before transaction writes. The existing transaction layer waits for outer commit. - Restore qualification deadlines from persisted bindings. Retry failed sweeps and unfinished assignments. - Add PostgreSQL tests for idle quiet, nested commit, rollback, deadlines, standby, drain, and failure recovery. Verify signals from actual Muse producers. Preserve startup, newly committed work, and failed sweep retries across idle drain without SQL polling. - Document the remaining recovery cadence and sleep limitations. ## Verification - `pnpm -r typecheck`: passed before rebase. Server typechecking also passed after rebase and the drain fix. - `pnpm build`: passed before rebase. - Final change: 33 maintenance, delivery coordinator, and scheduler tests passed. Maintenance used a fresh disposable PostgreSQL 17 database. - Muse control-plane tests: 254 passed against the same disposable database. - `pnpm test:run`: attempted, then stopped after repeated embedded PostgreSQL startup failures from this host's semaphore limit. Native-session, Muse, and AI-connection suites could not create local clusters. The complete suite runs in CI. - `git diff --check`: passed. - Final commit `e106cbf20d`: all CI gates passed, including complete test suites, E2E, build, production bundle boot, public-install smoke test, and canary dry run. Greptile and Paperclip source reviews both rated 5/5, with no unresolved findings. The focused suites accept `PAPERCLIP_MUSE_MAINTENANCE_TEST_DATABASE_URL` and `PAPERCLIP_MUSE_CONTROL_TEST_DATABASE_URL`. Use a disposable test database; fixtures remove test data. ## Risks - Unfinished assignments retain five-second recovery. This change removes idle polling, not all Muse polling. - Signals are process-local. Independent database writers need explicit integration. Startup recovers persisted state. - Qualification deadlines depend on the running process. Existing durable Muse sleep blockers remain unchanged. - The receiver keeps its batching timer. An empty buffer already makes no database calls. - No schema migration or API change. ## Model Used OpenAI Codex, GPT-6 family, with code execution and repository tools. The exact model version 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 (focused suites; full local run has the environment limitation 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> |
||
|
Devin FoleyandPaperclip
|
63fbb714e6 |
fix(server): release idle tracking after aborted static UI files (#16685)
## Thinking Path > - Paperclip runs work for AI agents. > - Hosts must prove that accepted work has finished before idle sleep. > - HTTP request tracking keeps unfinished writes active after a client leaves. > - Static file middleware has no returned completion promise. > - An aborted UI download can leave its request active forever. > - This PR completes only a matched file response and keeps other work tracked. ## Linked Issues or Issue Description Refs #15522. Found while qualifying #16679. A search found no duplicate static-download idle fix. **What happened?** Closing a client during a static UI download leaves one tracked request active. The response has closed, but the file middleware never calls the wrapped response end method. The instance cannot report idle after that request. **Expected behavior** A closed, matched static file response must not leave a permanent work hold. Unfinished application writes must remain tracked. A missing file must keep the normal fallback and error paths. **Steps to reproduce** 1. Serve a large local file through the static UI middleware with idle admission enabled. 2. Start the download over HTTP. 3. Disconnect the client after its first data chunk. 4. Observe that the tracked request remains active with the original middleware. **Paperclip version or commit** Reproduced on the behavior in `d30648dee89720358271c276b9fc28cf83f18be7`. This branch starts at `cfa9dce295970a85b3d8ee14cab0eccf50bc72e2`. ## What Changed - Mark the completion of a known read-only response without dropping pending middleware promises. - Wrap static UI file serving. A successful file match is required before a closed response can complete its tracking. A late match rechecks an already closed response. Missing files keep the normal fallback. - Use the wrapper for hashed assets, other static UI files, and public development files. Preserve caller headers and cache options. - Test both HTTP abort timings, nested routers, conservative mounted-application holds, parent writes, mutating GET fallbacks, and callback-only work. - Update the idle contract and its verification map. Correct the paused-message recipe from hosted qualification: an explicit Pause cancels older queued work, so test new accepted input after the pause and await the normal resume lifecycle. ## Verification - The original two real stream-abort regressions fail with the original Express static behavior. - Three review regressions fail on `e260f4694afab8447b0cb203f439cb3ccad254e8`: a client closing before file lookup, an unfinished parent in a mounted application, and an unfinished parent in a root without installed handler tracking. Fixed sources were restored byte for byte after the comparison. - Two focused suites: 16 passed. They cover nested routers, both abort timings, hidden parent writes, late missing-file fallbacks, contents, HEAD, conditional responses, and caller headers. - Module and map checks pass. Map checks are structural only. - `pnpm -r typecheck` and `pnpm build`: passed on `f242ee739344484995d3addf81635512370778c8`. - Full CI matrix: 54 successful checks and 2 intentional Storybook skips. - Apex: 5/5 on the exact current head, with zero unresolved threads. - Full local maintained test groups are still running before final handoff. - The initial local broad run was stopped before changing source for the P1 review findings. Two file-mode fixture failures in that run came from the scratch runner's inherited `0077` mask; the same suite passes with the normal child test mask `0022`, without source or host configuration changes. The replacement run uses `0022` for test children and command-scoped `init.defaultBranch=master` for existing Git fixture assumptions. - This PR has not been deployed to a hosted environment. ## Risks Only a matched file served by the admitted application with installed handler tracking can complete this way. Routers share that scope. Mounted Express applications and roots without handler tracking keep the existing conservative response hold; this change does not claim their hidden writes are tracked. An unfinished tracked parent promise still blocks sleep after a file client disconnects. Redirects, missing files, application callbacks, and other GET requests retain their guards. No authentication, company scope, provider cleanup, or scheduling gate is removed. No schema or API response changes. ## Model Used OpenAI GPT-6 through Codex, with reasoning, source editing, and shell tools. The session does not expose a more specific serving model ID or its 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> |
||
|
Devin FoleyandPaperclip
|
4415665d67 |
fix(server): allow proved idle waits and verify historical cleanup (#16633)
## Thinking Path > - Paperclip manages agents, task continuations, integrations, and their execution resources. > - Automatic idle sleep needs durable evidence that accepted work and provider cleanup are settled. > - Response-driven questions, passive board escalations, and completed connection receipts currently look like unfinished work. > - Legacy ephemeral leases and terminal image sessions need actual deletion proof before their history can become idle. > - This PR adds narrow evidence-based exceptions and read-only provider verification while preserving queues, outcomes, and ownership fences. > - Idle instances can sleep without losing responses, automatic reconciliation, or cleanup obligations. ## Linked Issues or Issue Description **What happened?** Instances with no executable work can remain awake because pending human responses, passive recovery records, outbound integration configuration, or completed delivery history are counted as background work. Older failed ephemeral leases and finished setup sessions also lack the evidence needed to distinguish retained resources from deleted allocations. **Expected behavior** Permit idle sleep only for proved response-driven waits and settled history. Continue blocking real queued work, inbound integrations without a wake contract, unfinished accounting/retries, retained resources, and uncertain provider cleanup. **Steps to reproduce** Finish an issue's producing run, leave a pending question or confirmation, and request an owned scheduled idle report. Separately retain a processed connection receipt or a failed ephemeral lease whose provider allocation has already been deleted. The old checks keep those instances awake even after executable work has finished. Related public groundwork: #15522, #15599, #15761, #15897, #15907, and #16325. Searched existing public idle-sleep PRs; this addresses additional blockers in the current predicates. ## What Changed - Recognize supported response-driven waits and passive board recoveries without changing task state, response audiences, or continuation policies. Bind passive issue generation into routine-only alarms too. Keep automatically polled merge-confirmation cards awake. - Treat processed events, delivered intents, disabled drafts, and outbound MCP configuration as history/configuration where proved safe. Preserve pending deliveries, downstream wake requests, and inbound GitHub bindings. - Permit only pristine Slack setup drafts with a matching disabled channel connection, no provisioning registration, and no prior provider identity/activity. A saved slash-command name remains inert configuration. - Repair missed routine execution-issue status hooks in bounded awake-only maintenance. Lock and re-read exact owners, apply existing completion/transient-failure rules, and preserve normal continuation after an issue resumes. - Allow human-only connection setup to wait only when no reusable active connection of its purpose exists; available access reblocks sleep for reconciliation. The delivery worker shares this proof, releases its local idle hold, and retains maintenance only while already awake. Passive-only ticks pause before admission during an owned drain, allowing slow backups to finish; new producer notifications still reconcile and invalidate the backup. - Add optional `environmentInspectLease` SDK discovery/dispatch and read-only Daytona/fake-provider implementations. Old workers remain conservative. Daytona status inspection uses a fresh provider GET without reconnecting or poisoning unused-worker eligibility. - Bind historical credential reads to an unrotated initial secret version and its immutable resolved-value hash; reject ambiguous histories and ambient Daytona credential/API-endpoint fallbacks before querying provider state. - Verify bounded rotating batches of historical resources on already-awake Cloud instances, outside readiness. Store host-owned deletion evidence bound to exact row identity/version and reject concurrent ownership changes. Preserve failure outcomes, cleanup/export intents, and run continuation authority. - Preserve cancellation provenance, defer contended routine rows individually, and cap actual inspection RPCs across caller timeouts and worker exits. Add database, SDK, provider, configuration-scope, and race regression coverage and update idle protocol documentation. ## Verification - `pnpm -r typecheck` and `pnpm build` passed on `24fd1c0f5ec`; server typecheck/build passed again after the final coordinator-only fix in `60f23e6e4b6`. - Final server regressions: 258 sleep-safety/connection cases and all 17 coordinator cases passed. Earlier focused checks also covered 331 combined idle/cleanup cases, 34 custom-image cases, 39 credential/cleanup cases, 39 connection-history cases, and 80 existing routine cases. Counts overlap and are not additive. - SDK: 153 passed. Daytona: 406 passed with 14 existing opt-in skips. The provider implementation also typechecks against the locked SDK used by the Cloud wrapper. - Dependency boundaries were exercised with real child workers holding provider inspection replies beyond caller timeouts, a real PostgreSQL row lock delaying one routine owner, and real connection creation → queue notification → worker hold release → resumption after access is enabled without another notification. A controlled six-second backup goes through the real coordinator, task drain, work signals, and idle-safety generation checks; maintenance permits completion, while a new committed transaction invalidates it. - Full local `pnpm test:run` completed its general-server phase with 21,122 passing, 29 failing, and 109 skipped tests; later phases did not run. That long run spanned source edits and is not a final-revision pass. The affected Slack draft assertion passes on current source. Restoring `/usr/sbin` for `lsof` produced 233 passing tests and two Git-fixture failures across the other five files (3 existing skips). Both remaining assertions pass with command-scoped `init.defaultBranch=master`; this host defaults to main and leaves both branches in their fixture. No source/global Git setting changed for that comparison. Original failure logs are retained. - A parallel typecheck/build attempt raced generated Runner outputs and failed an import. Running the build sequentially passed with no source change. - Local database checks use the maintained test-constructor seam with native PostgreSQL 16.14 because bundled PostgreSQL 18 is unreliable on this host. No database assertions were skipped in the focused runs. An interrupted combined run was superseded by the separate passing coordinator and safety/connection runs; no owned test database processes remain running. - Apex reviewed final head `60f23e6e4b615a08591e19d86443a6c35c091ce1`: **5/5**, zero unresolved threads. All current-head CI gates are green, with expected policy skips. The branch is conflict-free and GitHub reports a clean merge state. - Current-head CI initially failed `plan-task-fixture.spec.ts:66` during fixture deletion: the final company DELETE hit `activity_log_company_id_companies_id_fk` after the feature assertions completed. The deletion path is unchanged here; the final diff only changes drain-gated passive maintenance. The preceding revision passed this shard. Failure evidence is retained in [the original CI job](https://github.com/paperclipai/paperclip/actions/runs/39006936642/job/117321244396); the targeted retry passed on the unchanged commit ([retry job](https://github.com/paperclipai/paperclip/actions/runs/39006936642/job/117322535013)). This shows recovery, not that the underlying company-deletion race is resolved. - These checks do not claim staging/production deployment or live provider cleanup proof. Provider tests simulate responses. Full Cloud staging verification requires a compatible Core/private-plugin composition after the Core revision is on master. ## Risks Sleep eligibility changes require the existing owned drain, final generation validation, local-work checks, and plugin drain. Unsupported states fail closed. Queued agent input and routine execution remain blocked. Outbound MCP exemptions exclude inbound GitHub grants; other unsupported inbound integrations remain awake. The inspection RPC is optional and requires a matching live worker method. Its actual outstanding calls are capped at two across workers even after the caller times out. Historical lease inspection requires its recorded plugin/config; setup inspection requires an unchanged recorded config fingerprint. Not-found under replacement account credentials is not historical deletion proof: ambiguous/rotated secret histories defer verification, and initial-version reads must match their recorded value hashes. Missing provenance (including legacy image sessions without a recorded config fingerprint), live/stopped resources, provider authorization failures, and concurrent changes defer verification. Routine repair replays existing completion/transient-failure behavior and timestamps at reconciliation time; it does not execute or discard deferred comments. No migration or destructive provider operation is added. The new marker does not grant run-resume authority. Full Cloud staging validation requires a compatible Core/private-plugin composition; no production deployment is included. ## Model Used OpenAI GPT-6 Codex, with repository/tool access and code execution. Exact model identifier 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 relevant tests locally; passing focused checks and full-run limitations are documented above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> |
||
|
Devin FoleyandPaperclip
|
35b3345033 |
fix(server): support issue monitor deadlines in idle sleep reports (#16325)
## Thinking Path > - Paperclip manages work for companies and AI agents. > - Hosted workspaces can sleep when their required work has a safe wake path. > - The current idle check treats every open issue as background work. > - Backlog, unassigned and human-waiting records can therefore keep an idle process running. > - Future automated monitors still need their saved check time and execution claims. > - This PR adds a signed host capability for passive issues and verified monitor deadlines. > - The host can save one alarm for the earliest required routine or monitor time. ## Linked Issues or Issue Description **What happened?** A quiet workspace with a backlog or unassigned todo issue reports background work as present. A future issue monitor also prevents sleep even when the host can save a durable wake alarm. **Expected behavior** Proven passive records can sleep. A validated future monitor can sleep only after the host saves its alarm and validates the same owned report. Active, unknown and unresolved work stays awake. **Steps to reproduce** 1. Create a quiet workspace with an unassigned todo issue and no other work. 2. Acquire a bounded, signed Cloud idle hold. 3. Request the routine-only safety report with `scheduledWake=1`. The issue blocks sleep. 4. Request the new capability with `scheduledWake=2`. A passive issue can now report no work. 5. Add a valid future monitor to an assigned in-progress issue. The second capability returns its saved deadline for host acknowledgement. **Paperclip version or commit** Base: `c30fb7d2b75ea30f72d63656b2a7cfff7fe0929d`. **Deployment mode** Cloud-managed authenticated instance. Local scheduling remains in the existing scheduler. Related protocol: [#16215](https://github.com/paperclipai/paperclip/pull/16215). Open scheduler and monitor PRs were checked for overlap. This change leaves monitor admission, dispatch and quota recovery with their existing owners. The roadmap already includes scheduled work; this is a focused idle-safety extension. ## What Changed - Add signed scheduling capability `2`; capability `1` retains the routine-only rules. - Allow untouched onboarding prompts with idle assignees to sleep using the same awaiting-user predicate as recovery; typing, answering the opening card or a prior run removes the exemption. - Inspect passive issue state and future monitors in the same bounded, read-only snapshot as other work. - Check ownership, company scope, known states, claims, policy/deadline agreement, attempts, timeout and dependency repair. - Combine routine and monitor deadlines into a known proof kind and catalog generation. - Preserve other sleep blockers, including task watchdogs, conversations, live runs, external credentials and recovery. - Update the protected route, OpenAPI contract, documentation and PostgreSQL regression tests. ## Verification - Full workspace typecheck and build passed. Final server typecheck and full build passed after the final guards. - Initial safety, scheduling, protected-route and control-authentication checks: 326 tests passed. - Final safety and heartbeat scheduling checks: 253 tests passed. Includes reporting a future deadline, restoring scheduling from PostgreSQL and one dispatch across competing schedulers. - Local general server checks: 20,260 passed; UI: 8,460 passed; CLI: 534 passed, including isolated reruns of two embedded-PostgreSQL startup failures. All remaining workspace groups and 158 serialized suites passed. The suite was completed in groups after the isolated infrastructure retries; the final onboarding change also passed the focused 674-test recovery/safety run. Earlier unchanged Git fixtures failed under Xcode's system `init.defaultBranch=main`, reproduced on the clean base and passed with a test-scoped fixture default; no source or machine configuration was changed. - Rebased issue, protected-route and scheduler checks: 115 tests passed; server typecheck passed. - The expanded change passed 674 idle-safety, scheduling and recovery regression tests, including six onboarding admission/invalidation cases. Final server typecheck and build passed. - Current-head Apex is 5/5 with no unresolved threads. All 54 current-head checks passed, with two expected skips. Four jobs were interrupted by runner shutdowns or lost communication on their first attempts; their targeted reruns passed. Live hosting qualification awaits the verified master release and matching host capability. ## Risks This intentionally covers a limited set of passive states and valid future monitors. Assigned active tasks without a proved waiting path, unknown states, claimed monitors and quota recovery still block sleep. It does not make every open issue safe to sleep. A matching host must recognize `routines_and_issue_monitors`, acknowledge the durable alarm and revalidate the exact owner, generation and deadline before stopping. The report alone grants no stop authority. Older hosts retain capability `1`. Independent database writers remain outside the admission contract. Existing scheduler claims and execution authority are unchanged. No schema migration is required. Roll back host negotiation to capability `1` to restore conservative admission, while retaining alarm delivery until outstanding alarms have been handled. ## Model Used - OpenAI GPT-6 (Codex), with repository editing, command execution and GitHub tools. The exact deployment ID and context-window metadata 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> |
||
|
DottaandPaperclip
|
7c016c049b |
perf(workspaces): wake terminal cleanup from committed changes (#16282)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Agents use execution workspaces to retain their files and Git branches. > - A reaper closes eligible workspaces after their task tree is terminal. > - The reaper scanned every heartbeat tick and read Git before cheap eligibility checks. > - Permanent scans also query idle databases when no cleanup work exists. > - This pull request wakes cleanup after transaction settlement and restores bounded work at startup. > - The benefit is fewer scans while fresh cleanup guards continue to protect retained work. ## Linked Issues or Issue Description Related work: Refs #13619, Refs #11753. Those PRs bound scans or back off unchanged candidates. This PR uses the transaction work coordinator to remove the permanent reaper tick. It overlaps their early terminal-state filtering but does not change their runtime restart behavior or introduce a second Git scheduler cache. **What existing behavior does this improve?** Terminal workspace cleanup and Git inspection for workspace detail reads. **Subsystem affected** Server orchestration, execution workspaces, and idle sleep admission. **Current behavior** The heartbeat tick scans workspace candidates every thirty seconds. It runs Git before rejecting nonterminal trees, cooldowns, and active runs. Display inspections expire after five seconds. **Proposed behavior** Recover workspace rows in bounded startup pages. Wake after relevant issue, workspace, and pull-request work-product transactions settle. Schedule the terminal tree cooldown or a five-minute eligibility retry. Back off unchanged candidates across unrelated wakes, and invalidate their backoff after relevant committed changes. Park empty and nonterminal-only lanes. Reuse display inspections for thirty seconds. Read fresh Git before destructive cleanup. **Reason and benefit** Avoid recurring database and Git work when no cleanup is possible. Retain transaction admission holds, restart recovery, and fresh deletion checks. Retention eligibility can wait for the next host wake; active and failed recovery work still blocks sleep. **Breaking changes** No schema or API shape changes. Display Git state can lag by thirty seconds. Cleanup eligibility retries can take five minutes. ## What Changed - Add a terminal-workspace worker to the existing transaction work coordinator and scheduler. - Signal relevant writes before mutation on the owning transaction. Outer transaction settlement wakes cleanup. - Derive cooldown deadlines from the full company-scoped terminal task tree. Continue bounded recovery pages without a permanent sweep. - Bound unchanged candidate inspections to five minutes, retain at most 256 entries, and invalidate affected entries with scoped transaction hints. Repeated wakes preserve the earliest inspection deadline. - Collect deadlines from task trees already read by bounded pages. Restore the cursor after a failed page and rescan commits received during a rotation. - Check terminal state, cooldown, and active runs before Git. Preserve lifecycle locks, generation fences, and fresh final checks. - Keep failed or incomplete recovery visible to idle admission. Preserve the passive retention exception. - Extend the bounded display cache to thirty seconds and document the scheduling and wake limits. ## Verification - Focused workspace, coordinator, cache, and local sleep suites: 111 tests passed with real local PostgreSQL and Git worktrees where applicable. - Final worker, display cache, and local sleep suites: 28 tests passed. - The regressions prove three ineligible workspaces invoke zero Git scans, cached clean display state cannot remove new untracked files, and real producer notifications wait for outer transaction settlement. - `pnpm -r typecheck` and `pnpm build` passed locally. Map and module-boundary checks passed. - Work-product and sleep-safety suites: 221 tests passed. Reopen and issue-tree control suites: 46 tests passed. Final backoff, workspace, worker, and work-product suites: 87 tests passed. - Final workspace and worker suites: 81 tests passed. Deadline selection performs zero database reads. A failed full page retries its unread terminal workspace before recovery can park. - Full CI coverage passed on head `2c397970f7`: general tests, serialized server suites, runner verification, E2E shards, typecheck, build, and the final verification gate. The duplicate serial local `pnpm test:run` was stopped after full CI coverage passed; it is not reported as a completed local pass. - Greptile: 5/5 with both findings resolved. The security scan also passed. No failing or pending PR checks remain. ## Risks - Display state can be stale for thirty seconds. Destructive guards bypass the display cache. - Undelivered or otherwise ineligible terminal work retries after five minutes instead of each heartbeat tick. - Transaction signals cover the owning `createDb` scope. Direct SQL and independent server processes require startup recovery. - Local cooldown timers are maintenance hints. They do not establish an external wake alarm or whole-instance sleep safety. Existing blockers remain in force; failed bounded recovery keeps the host awake. ## Model Used OpenAI GPT-6, model ID `gpt-6`, through Codex. Reasoning, repository inspection, shell execution, and browser tools assisted the change. The runtime does not expose the 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> |
||
|
Devin FoleyandPaperclip
|
de335b42d4 |
Expose signed routine wake deadlines during idle sleep checks
Expose signed routine wake deadlines during idle sleep checks Report the earliest eligible routine deadline to an authenticated host scheduler, while preserving local execution, owned sleep leases and conservative blocking for unsupported work. Verified with focused tests, typecheck, build, all required CI checks and Apex 5/5. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
Devin FoleyandPaperclip
|
d49cbf4c4f |
Permit idle sleep with consumed retries and passive worktrees (#15907)
## Thinking Path
> - Paperclip is the open source app people use to manage AI agents for
work.
> - Idle sleep must wait for accepted work and durable cleanup.
> - The retry dispatcher keeps its due date after it promotes a retry.
> - Finished retries and passive local directories can therefore look
like pending work.
> - This pull request requires completion proof and permits known
passive directory records.
> - It retains the records and keeps real work and unknown state awake.
## Linked Issues or Issue Description
Related: #15897 and #15851. The open runtime-service work in #13413
retains its separate service and ingress guarantees. This patch does not
change those checks.
**What happened?**
Idle sleep treats a consumed retry date as pending work. It also blocks
on every retained execution workspace, including a local directory
without runtime activity.
**Steps to reproduce**
1. Promote a scheduled retry, then finish its run without accounting
debt. Retain its scheduled retry date.
2. Retain a local primary directory or Git worktree without runtime
configuration, commands, or cleanup due.
3. Request an owned idle safety report. The report returns `present` for
this history.
**Paperclip version or commit**
Reproduced on master at
|
||
|
Devin FoleyandPaperclip
|
4b59e2a2ec |
fix(server): allow idle sleep with saved AI auth and verified lease history (#15897)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - Automatic idle sleep must preserve accepted work, cleanup and recovery. > - Saved AI login connections are configuration without independent workers. > - Failed ephemeral leases can remain after their resources have stopped. > - Detached subscription observations and quota refreshes can outlive a response. > - This pull request distinguishes inert configuration and verified history from unfinished work. > - The benefit is safe idle eligibility while keeping credentials and audit history. ## Linked Issues or Issue Description **What happened?** An active saved AI login blocks idle sleep after its login and agent runs finish. A failed ephemeral lease also blocks after successful cleanup and provider termination. The connection guard currently hides detached quota and subscription work that can outlive the request that started it. **Expected behavior** An idle instance can sleep with saved AI credentials and verified failed-lease history. Accepted refresh work, recovery journals, unfinished login sessions, pending cleanup and run accounting must still block. **Steps to reproduce** 1. Save an AI connection with purpose `ai`, transport `runtime_auth` and status `active`. 2. Finish login and all agent work, then acquire an owned idle drain. 3. Read the idle safety report. The saved connection alone reports background work. 4. Repeat with a failed ephemeral lease whose exact provider termination and finished run are recorded. Its history also reports background work. **Paperclip version or commit** Based on `0bbd586a2f6`, after updates from master. **Deployment mode** Single-process managed or self-hosted server using the owned idle-drain protocol. Related work: #15522, #15761 and #15851. Searched open and closed pull requests for related idle-sleep changes. The open accounting recovery work in #15879 is separate; this change does not settle or discard accounting debt. ## What Changed - Exempt active saved AI runtime-auth configuration from the connection sleep check. Other connection types and unknown states keep their existing checks. - Track detached subscription discovery and provider observations until their actual promises settle. - Track underlying quota work through an HTTP timeout, including an OAuth exchange and the save of its replacement credential. - Inspect the credential-recovery directory without reading credential bytes. Journals, probes, temporary files, unexpected entries and read failures keep sleep blocked. Empty company directories left by successful recovery permit sleep. - Exempt a failed ephemeral lease only after successful cleanup, a release timestamp, an exact company/run/lease/resource termination receipt, and a finished run without retry or accounting debt. Export and cleanup intents still block. - Use the same failed-lease proof when checking finished login-session cleanup. - Add PostgreSQL and filesystem regression tests and document the behavior. ## Verification - Four focused suites passed: 186 tests covering durable guard decisions, local recovery, detached subscription work and quota timeouts. - `pnpm build` passed. - `pnpm -r typecheck` passed. - `pnpm --filter @paperclipai/server build` and a final server TypeScript check passed after the directory-scanner cleanup. - The final directory-scan recheck passed all 18 tests. - The original-base `pnpm test:run` attempt found three local failures. It was stopped before completion to update the branch after two master changes. All three also reproduce on unchanged master `e733bddc45e`: two Git default-branch fixtures in `workspace-runtime.test.ts` and one native runner recovery fixture in `native-session-resume.test.ts`. - Both Git fixtures pass with a command-scoped `init.defaultBranch=master` setting. The native fixture fails with the same result on both branches. No global Git setting was changed. - All CI gates passed on the final head `ff63497d9bef40f5f00567bc729fb17536cfe136`: [complete CI run](https://github.com/paperclipai/paperclip/actions/runs/38156312722). The only skipped checks are the two path-gated Storybook jobs. - Greptile Apex scored the final head 5/5 with no actionable findings and no unresolved review threads. The branch has no merge conflicts with master. - After the master update, `pnpm build` and `pnpm -r typecheck` passed again. All four focused suites passed again: 186 tests. - Staging deployment and live sleep/wake verification are pending the Core merge and the verified private Cloud build. No production rollout has been started. ## Risks - The failed-lease exception is deliberately narrow. Older records without exact termination evidence or a complete run receipt continue to block. - Pending OAuth recovery blocks sleep until it is recovered. No credential, history, cleanup obligation or accounting receipt is deleted. - The existing owned drain, ingress fence, local-work generation checks, plugin drain and backup checkpoint remain required. - No schema migration or provider mutation is introduced. Roll back to the previous application release to restore the conservative guard behavior. ## Model Used OpenAI GPT-6 (Codex), with repository inspection, code editing, test execution and read-only provider analysis. No additional model version or context-window size was supplied by the runtime. ## Checklist - [x] I have included a thinking path that traces from project context to this change - [x] I have specified the model used (with version and capability details) - [x] I have checked ROADMAP.md and confirmed this PR does not duplicate planned core work - [x] I have searched GitHub for duplicate or related PRs and linked them above - [x] I have either linked existing issues or 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 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 risks - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no unresolved comments - [x] I will address all reviewer comments before requesting merge Co-authored-by: Paperclip <noreply@paperclip.ing> |
||
|
Devin FoleyandPaperclip
|
b36960b758 |
fix(server): permit verified question-delivery history to sleep (#15851)
Treat completed question-response deliveries as history only after their exact same-company handoff finishes. Preserve missing evidence, retries, accounting debt, and idle-sleep fences. Add a concurrent all-state question history index with interrupted-build repair. Keep migrations serialized without waiting snapshots that deadlock concurrent index builds. Validated by 118 focused sleep tests, six migration/snapshot tests, full repository typecheck, all CI gates, and Apex 5/5. Co-Authored-By: Paperclip <noreply@paperclip.ing> |
||
|
Devin FoleyandPaperclip
|
baa94bbea5 |
Fix idle sleep checks for finished login and wake history (#15761)
## Thinking Path
> - Paperclip manages AI agent work and keeps a durable record of each
run.
> - Automatic idle sleep must wait for accepted work and provider
cleanup.
> - The current sleep check treats every saved login session as active
work.
> - It also treats every coalesced wake request as pending work.
> - These records can remain after their work and cleanup finish.
> - This pull request checks the cleanup and linked-run evidence before
it treats these records as history.
> - The change lets otherwise idle instances sleep and keeps uncertain
work awake.
## Linked Issues or Issue Description
**What happened?**
A finished login session blocks idle sleep after its provider resource
has been removed. A consumed coalesced wake also blocks sleep after its
linked run finishes.
**Expected behavior**
Finished history should permit sleep when cleanup or run completion is
confirmed. Active work, missing evidence, retry debt, and pending
accounting must still block sleep.
**Steps to reproduce**
1. Create a finished login session with a retained provider reference.
2. Record a successful release for the same resource, company, and
environment.
3. Check idle sleep safety under an owned, quiet admission hold.
4. The current check reports background work. The new check treats the
session as history.
5. Repeat with a finished coalesced wake linked to a finished run in the
same company and agent.
**Paperclip version or commit**
Reproduced on master at
|
||
|
Devin FoleyandPaperclip
|
9b624a110e |
fix: checkpoint database backups before idle sleep (#15629)
## Thinking Path > - Paperclip coordinates agent work and preserves company state. > - Hosted instances can stop compute when admission and all work sources are quiet. > - The periodic database backup timer currently blocks every automatic sleep. > - Disabling backups would remove recovery points. > - This change offers an explicit final-backup checkpoint before sleep. > - A verified archive and restart marker preserve recovery while compute is stopped. ## Linked Issues or Issue Description **What happened?** An otherwise idle instance with automatic backups enabled always reports background work. Operators cannot reclaim idle compute without disabling those backups. The work inventory also queries the database before checking known local blockers. **Expected behavior** An operator can enable checkpoint mode. The server may authorize sleep only after all other work is quiet and a fresh backup is verified under the same owned hold. Backups resume on restart with the existing retention policy. **Steps to reproduce** Acquire a bounded idle drain on an instance with no application work and automatic backups enabled. Read its owned idle safety report. It reports present even when all other work is clear. With this change and `PAPERCLIP_DB_BACKUP_IDLE_CHECKPOINT_ENABLED=1`, a successful final backup can permit sleep; backup failures still refuse it. **Paperclip version or commit** Reproduced from master at |
||
|
Devin FoleyandPaperclip
|
3367b75ccc |
fix: fence accepted work and cleanup before idle sleep (#15522)
## Thinking Path > - Paperclip is the open source app people use to manage AI agents for work. > - A host can stop an idle instance to reduce unused compute. > - Zero active runs do not prove that requests, saved work, or cleanup are complete. > - A client can disconnect while its request still writes, and failed cleanup can remain only in memory or on disk. > - This pull request holds new admission and checks accepted work, cleanup, and persisted work under an owned drain. > - The host gets an empty report only while those checks remain valid. ## Linked Issues or Issue Description Refs #13413. That companion PR uses the same owned-hold vocabulary for runtime services. This PR covers HTTP requests, scheduler work, accounting, cleanup, and the instance work inventory. It does not include the preview gateway changes. **What existing behavior does this improve?** The instance task-drain API reports process counters. It does not prove that a host can safely stop the instance for idle sleep. **Current behavior** A quiet run set can coexist with an unfinished request, cleanup after a completed run, future work, or a failed accounting write. **Proposed behavior** Provide a bounded owned idle hold. Block new ingress, track accepted handler promises, inspect durable and local work, and return `none` only when the same hold stays quiet through the checks. Keep normal deployment drains compatible. **Reason and benefit** Hosts can identify eligible idle instances without treating a disconnect or failed cleanup write as completed work. ## What Changed - Add `purpose: "idle"`, a bounded TTL, unique owners, and owner-checked release to task drain. Existing holds cannot be replaced by another API request. - Gate HTTP ingress before parsers, auth, webhooks, and MCP. Gate new WebSocket upgrades. Track async handlers in nested Express routers and error middleware until they settle, even after the response or client disconnect. - Count accepted live-event WebSocket authentication through settlement, even after disconnects. Count detached built-in agent, managed-home and runtime-service startup reconciliation after readiness. - Keep health and control mutations tracked. Count control-request authentication separately from the read-only report, including concurrent user/company/membership writes. - Pause new scheduler admissions during idle holds. Count work already in flight, including database backups, and reject scans whose work generation changes. Periodic backups block sleep without a host wake schedule. - Inspect accounting and orphan-cleanup spool directories without skipping temporary or malformed entries. Retain orphan tokens through queue splices, flush failures, and buffer overflow. Keep failed usage capture counted until its database failure fence is written. - Check persisted work across companies in a bounded read-only transaction. Include deferred agent-file cleanup and saved watchdogs whose watched issues are complete. Enabled plugins and unsupported retained work remain blockers. - Require the exact idle owner and completed startup tracking before returning an empty report. Keep reports free of tenant details. - Document the hosting protocol, retry behavior, conservative blockers, and the remaining external provider-stop race. ## Verification - Current head: `8d9a599b732c2047b8c671d4799065d7c90a3567`, rebased on master `941a3fa991aeb97eb1ac390c65b7973b5f6de1ad`. Both heartbeat helper extractions are preserved. GitHub confirms no merge conflicts. - Focused heartbeat renderer/run-log, drain, control-auth, admission, route and PostgreSQL inventory checks: 234 tests passed in 10 suites after the rebase. - The POST task-drain contract includes the expected `409` conflict response. The 84 OpenAPI and instance-settings route tests and server typecheck passed after that final documentation fix. - Final accepted-upgrade/startup regression run: 104 tests passed in five suites, including success and failure after readiness or disconnect. Final server typecheck also passed. - `pnpm -r typecheck`: passed. - `pnpm build`: passed. - `pnpm install --frozen-lockfile` and the PR diff secret scan passed. No dependency or lockfile changes are added by this PR. - The prior verification covered HTTP disconnects and early responses, async error handlers, concurrent authentication writes, signed bootstrap, saved watchdogs, backup promises, orphan cleanup, spools and failed accounting fences. Those tests passed. - The last full local `pnpm test:run` stopped in the general-server phase with 16,658 passing tests and five skill/connector fixture failures caused by an ancestor workspace skill directory. That full local run preceded this rebase and has not been repeated for the import conflict. Full current-head CI passed: 53 successful checks and two conditional skips, with no failures. - All eight review findings are fixed and their threads resolved, including accepted upgrade authentication, detached startup writes and the POST conflict contract. Current-head Apex review is 5/5 with no new findings; all eight review threads remain resolved. - No live provider stop or production change was performed. ## Risks - The hosting controller must use the owned protocol and hold external admission through its final validation and provider stop. A legacy drain cannot authorize idle sleep. During an idle hold, new requests receive 503 with `Retry-After: 1`; the host must handle queueing or retry before enabling this path. - This is a single-process protocol. An unexpected restart after the last validation can race an external provider stop. The host must serialize deploy/wake/sleep operations and bind the validation to the instance it stops. Multiple replicas need shared fencing. - Some retained state conservatively prevents sleep, including every enabled plugin. This PR does not promise that every inactive instance becomes eligible. - The HTTP adapter uses Express 5 router layers. Real Express tests cover nested routes, errors and disconnects. New routes must register before tracking is installed. Detached work must have durable state or explicit work tracking. - Unrecoverable in-memory cleanup debt keeps the instance awake until reconciliation. This change does not make such debt survive an unplanned process crash. No schema migration or provider configuration change is included. ## Model Used OpenAI GPT-6 via Codex, with reasoning, repository inspection, tool use and code execution. The exact deployment variant/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 and contains no internal Paperclip ticket id or instance-derived details - [x] I have run focused tests locally and they pass; the full local fixture limitation and clean full CI result are documented above - [x] I have added or updated tests where applicable - [x] I have updated relevant documentation to reflect my changes - [x] I have considered and documented any risks above - [x] All Paperclip CI gates are green - [x] Greptile is 5/5 with no open P2s, recommendations, or follow-ups - [x] I will address all Greptile and reviewer comments before requesting merge --------- Co-authored-by: Paperclip <noreply@paperclip.ing> |