Repository navigation
chore(release): prepare 0.17.0 - #496
Conversation
|
🦞👀 Pull request received. I will update this pull request when review starts. ClawSweeper review completeClawSweeper finished reviewing this revision. The review result is being finalized. |
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
|
Codex review: needs maintainer review before merge. What this changesThis PR prepares 0.17.0, patches the development Glob dependency, and isolates Windows relay tests from process-query failures. Example: Preparing the next mcporter release
Review scores
ProductKind: Maintenance · Worth it: Yes · Fix scope: Complete Merge readiness✅ Ready for maintainer review Keep this maintainer-authored release PR open: its changes are absent from main, and no actionable patch defect was found. Priority: P2 Before mergeNone. FindingsNone. Agent review detailsHow this fits togetherRelease metadata identifies the published CLI and library, while the development cleanup chain uses rimraf and Glob; relay integration tests exercise the daemon with real MCP children. flowchart TD
A[Release metadata] --> B[Package and native artifacts]
B --> C[CLI and library users]
D[Development cleanup] --> E[rimraf]
E --> F[Glob 14.0.2]
G[Relay tests] --> H[Daemon and real MCP child]
Technical reviewBest possible solution: Land the release preparation after the existing branch qualification completes, then follow the documented signed-release gates. Do we have a high-confidence way to reproduce the issue? This is release preparation; the vulnerable Glob resolution is directly visible on main and matches the published advisory. Is this the best way to solve the issue? The version bump, preserved release entries, targeted dependency override, and existing test-isolation pattern form a focused release change. AGENTS.md: found and applied where relevant. Codex review notes: model internal, reasoning medium; reviewed against 50efe6ae7470. Provenance checked
TestingProof path: in-process harness. SecurityNone. EvidenceWhat I checked:
Preflight sourceSource unknown. Source comparison at the observed PR snapshot only. Completion is self-reported; body, policy, tests, execution, and report bytes are not verified. LabelsLabel changes:
Label justifications:
Rating scale6/6 🦀 challenger crab · 5/6 🦞 diamond lobster · 4/6 🐚 platinum hermit · 3/6 🦐 gold shrimp · 2/6 🦪 silver shellfish · 1/6 🧂 unranked krab. Overall follows the weaker of proof and patch quality; ✨ marks media proof (a screenshot, video, or linked artifact) that directly shows the changed behavior. WorkflowClawSweeper edits this one comment on every review. Comment Reviewed November 5, 2026, 1:56 PM ET / 18:56 UTC. |
Prepare mcporter 0.17.0. At the 256-transport limit, the daemon can reclaim its least recently used quiet transport before the idle deadline; active calls, Chrome owners, unknown outcomes, and failed retirements remain protected. This behavior change warrants a minor release.
Preserve all 19 Unreleased entries and contributor credits. Add a newly detected development dependency fix for Glob's cancellation-bypass advisory (GHSA-3g3m-2fjr-6pw4), overriding rimraf's Glob to 14.0.2. The patch has been available since October 27, satisfying the 48-hour cooldown, and retains Node 24 support.
Validation: release metadata and changelog bullet-preservation checks; independent P0–P2 review; source, installed npm package, native build, and release contract gates; synthetic async/sync rimraf glob cleanup. Release publication additionally requires the merged-head rehearsal and signed/notarized native verification.
Windows CI exposed a transient process-observation failure in the Chrome relay fixture. Match the existing discovery/elicitation test boundary: isolate this relay test from Windows WMI, while retaining its real MCP child and the dedicated process-retirement integration coverage. The 37 relay and process-retirement tests pass locally; static checks and updated independent review are clean.