fix(mcp): align linked-issue extraction with server after #4039

[RealDiligent]

Summary

Fixes a correctness bug where MCP local-branch analysis and the predicted-gate engine still used a stale linked-issue extractor after server-side fixes in #4039 (inline code spans) and #3862 (qualified owner/repo#N scoping).

Root cause

packages/gittensory-mcp/lib/local-branch.js extractLinkedIssues() scanned raw text without stripping markdown inline code. This repo's PR template checklist literally contains (e.g. Closes #123), so every local branch analysis that keeps that line spuriously reports linkedIssues: [123]. Those values are POSTed to /v1/local/branch-analysis and merged server-side, corrupting predicted-gate verdicts, collision signals, and duplicate detection.

The predicted-gate engine copy in packages/gittensory-engine/src/signals/predicted-gate-engine.ts had the same missing code-span strip, causing predictedGate and preflight to disagree inside one API response.

Fix

• Port the canonical repositories.ts extractor logic into MCP extractLinkedIssues(text, repoFullName)
• Pass repoFullName from collectLocalBranchMetadata
• Align predicted-gate-engine extractLinkedIssueNumbers with the same strip + unified regex
• Regression tests mirroring test/unit/db-parsers.test.ts template-checklist case

Impact

Local MCP users and miners running gittensory_predict_gate no longer get false linked-issue matches from template boilerplate; gate predictions align with server preflight.

Tracking issue

No open contributor-eligible upstream issue exists for this parity gap (maintainer PR #4039 merged without filing one; CreateIssue is denied for external contributors). Please create/associate a gittensor:bug tracking issue before merge per CONTRIBUTING policy.

Test plan

• npm run typecheck
• npm run build:mcp
• npx vitest run test/unit/local-branch.test.ts test/unit/predicted-gate-engine-branch-coverage.test.ts test/unit/db-parsers.test.ts

[superagent-security]

Superagent didn't find any vulnerabilities or security issues in this PR.

[loopover-orb]

Caution

🟥🟥🟥🟥🟥🟥🟥🟥🟥🟥🟥🟥

🛑 Gittensory review result - reject/close recommended

_{Review updated: 2026-07-07 16:20:06 UTC}

4 files · 1 AI reviewer · 2 blockers · readiness 100/100 · CI pending · blocked

🛑 Suggested Action - Reject/Close

• No linked issue detected — If this PR is intended to solve an issue, link it explicitly in the PR body.
• AI reviewers agree on a likely critical defect: No linked/associated eligible open issue: the PR description itself states none exists and explicitly asks a maintainer to file a `gittensor:bug` tracking issue before merge, which this repo's contributor-PR policy requires regardless of fix quality. — Resolve the flagged defect, or override if the AI reviewers are mistaken, then re-run the gate.

Review summary
This ports the canonical server-side linked-issue extractor (inline code-span stripping + owner/repo#N scoping from #4039/#3862) into the MCP local-branch analyzer and the predicted-gate engine, closing a real divergence where this repo's own PR template checklist text `(e.g. \`Closes #123\`)` was being parsed as a genuine issue link. The fix lands at the correct layer (all three extractors now share identical strip+regex+dedup logic) and the new tests exercise the actual failing input — the literal template line — rather than a fabricated payload, so this is real coverage, not theater. The PR's own description states no eligible tracking issue exists for this repo and asks a maintainer to create one before merge, which is the primary open item.

Blockers

• No linked/associated eligible open issue: the PR description itself states none exists and explicitly asks a maintainer to file a `gittensor:bug` tracking issue before merge, which this repo's contributor-PR policy requires regardless of fix quality.

Nits — 5 non-blocking

• packages/gittensory-mcp/lib/local-branch.js: the diff only shows the call-site line passing `repoFullName` into `extractLinkedIssues(...)`; confirm `repoFullName` is actually in scope in `collectLocalBranchMetadata` (not shown in the truncated diff) since the function previously didn't reference it.
• packages/gittensory-mcp/lib/local-branch.js: `extractLinkedIssues` now dedupes internally via a `seen` Set, but its only call site already wraps the result in `[...new Set([...])]` — the internal dedup is redundant, not wrong.
• The intentional behavior change of dropping keyword-less bare `feat(data): add contributor reconciliation reports #123` matching in the MCP extractor is a user-visible change for any local caller relying on old behavior; worth a one-line CHANGELOG/README note since MCP is also consumed by external tooling, not just this repo's server.
• Attach or create the tracking issue referenced in the PR description before merge, per repo policy.
• Add a brief comment or test at the `local-branch.js` call site confirming `repoFullName`'s source (e.g. parsed from git remote) so future readers don't need to trace it themselves.

Why this is blocked

• No linked/associated eligible open issue: the PR description itself states none exists and explicitly asks a maintainer to file a `gittensor:bug` tracking issue before merge, which this repo's contributor-PR policy requires regardless of fix quality.
• No linked issue detected — If this PR is intended to solve an issue, link it explicitly in the PR body.

Signal	Result	Evidence
Code review	❌ 2 blockers	1 reviewer
Linked issue	✅ Linked	#123
Related work	✅ No active overlap found	No same-issue or scoped active PR overlap found.
Change scope	✅ 20/20	Low review scope from cached public metadata (1 linked issue).
Validation posture	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 248 registered-repo PR(s), 73 merged, 0 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	RealDiligent; Gittensor profile; 248 PR(s), 0 issue(s).
Gate result	❌ Blocking	Repo-configured hard blocker found.

Review context

• Author: RealDiligent
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: JavaScript, Ruby, Svelte, TypeScript, Cuda, Markdown
• Official Gittensor activity: 248 PR(s), 0 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Keep the PR focused and include validation evidence before maintainer review.

Signal definitions

• Related work = same linked issue, overlapping active PRs, or title/path similarity.
• Change scope = cached public metadata such as size labels, draft state, and review-burden hints.
• Validation posture = whether the PR provides enough public validation/test evidence for maintainer review.
• Contributor workload = public contributor activity and cleanup pressure, not a repo-wide quality failure.
• Contributor context = public GitHub/Gittensor identity context; non-Gittensor status is not a blocker.

_{🟩 Safe / merged · 🟦 Advisory · 🟨 Held for review · 🟥 Blocked / closed}

---

💰 Earn for open-source contributions like this. Gittensor lets GitHub contributors earn for the work they already do — register to start earning →.

Checked by Gittensory, a quiet PR intelligence layer for OSS maintainers.

• Re-run Gittensory review

[loopover-orb]

Gittensory is closing this pull request on the maintainer's behalf (No linked issue detected; AI reviewers agree on a likely critical defect: No linked/associated eligible open issue: the PR description itself states none exists and explicitly asks a maintainer to file a `gittensor:bug` tracking issue before merge, which this repo's contributor-PR policy requires regardless of fix quality.; duplicate of another open PR). This is an automated maintenance action — to pursue this change, please open a new pull request with the issues resolved. Closed PRs may be analyzed later to improve review accuracy, but they are not automatically reopened or re-reviewed.

[codecov]

Codecov Report

❌ Patch coverage is 90.90909% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 93.58%. Comparing base (0979321) to head (cd4f08f).
⚠️ Report is 1 commits behind head on main.

Files with missing lines	Patch %	Lines
...ensory-engine/src/signals/predicted-gate-engine.ts	90.90%	0 Missing and 1 partial ⚠️

❌ Your patch status has failed because the patch coverage (90.90%) is below the target coverage (99.00%). You can increase the patch coverage or adjust the target coverage.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #4054      +/-   ##
==========================================
- Coverage   93.59%   93.58%   -0.01%     
==========================================
  Files         379      379              
  Lines       35575    35582       +7     
  Branches    13044    13045       +1     
==========================================
+ Hits        33295    33301       +6     
  Misses       1618     1618              
- Partials      662      663       +1     

Files with missing lines	Coverage Δ
...ensory-engine/src/signals/predicted-gate-engine.ts	99.67% <90.90%> (-0.33%)	⬇️

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.