feat(engine): extract computeLocalScorerTokens to gittensory-engine

[reyanthony062001-ops]

What

computeLocalScorerTokens (#782) is a pure ~35-line token scorer over changed-file metadata (paths + line counts, never source content), documented as needing to run "in every surface (stdio package AND hosted Worker)" — but it lived in src/signals/local-scorer.ts where the published gittensory-mcp/gittensory-miner CLIs cannot import it. This extracts it to packages/gittensory-engine/src/local-scorer.ts, with src/signals/local-scorer.ts becoming a thin re-export shim — the same layout as the #2278 duplicate-winner, #2282 scoring, and #4254 issue-rag-query extractions.

Its only dependencies are the already-portable isCodeFile/isTestPath (engine signals/test-evidence.ts) plus 3 narrow type shapes. Per the issue's guidance ("Lift or narrowly duplicate ... do not attempt to move local-branch.ts itself"), those 3 shapes are narrowly duplicated into the engine module (structurally identical to local-branch.ts's), leaving the large, Node-coupled local-branch.ts untouched in the backend.

Deliverables

• computeLocalScorerTokens extracted to packages/gittensory-engine/src/local-scorer.ts
• The 3 dependent type shapes narrowly duplicated into the engine module (not moving local-branch.ts)
• src/signals/local-scorer.ts is a thin re-export shim (relative source path, matching prior extraction shims)
• Exported from the package public entrypoint (index.ts)
• Existing tests pass unmodified: test/unit/local-scorer.test.ts (+ local-scorer-adapter + mcp-run-local-scorer) green through the shim; plus a new engine-side barrel test

Validation

• npm run test --workspace @jsonbored/gittensory-engine (build + 277 tests, 0 fail)
• npm run typecheck (0 errors)
• npx vitest run test/unit/local-scorer.test.ts test/unit/local-scorer-adapter.test.ts test/unit/mcp-run-local-scorer.test.ts (15 tests)
• npm run test:engine-parity
• git diff --check

Closes #4253

[superagent-security]

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 93.95%. Comparing base (c823f9c) to head (df0ce0f).
⚠️ Report is 44 commits behind head on main.
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4371   +/-   ##
=======================================
  Coverage   93.95%   93.95%           
=======================================
  Files         400      400           
  Lines       36817    36817           
  Branches    13451    13451           
=======================================
  Hits        34592    34592           
  Misses       1569     1569           
  Partials      656      656           

Files with missing lines	Coverage Δ
packages/gittensory-engine/src/local-scorer.ts	100.00% <100.00%> (ø)

🚀 New features to boost your workflow:

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

[loopover-orb]

Tip

🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩

✅ Gittensory review result - approve/merge recommended

_{Review updated: 2026-07-09 17:15:42 UTC}

4 files · 1 AI reviewer · no blockers · readiness 100/100 · CI green · clean

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
Clean, faithful extraction of the pure computeLocalScorerTokens token scorer into packages/gittensory-engine, matching the established shim pattern from #2278/#2282/#4254. The logic itself is byte-identical to the pre-diff src/signals/local-scorer.ts implementation, and the three dependent type shapes are narrowly duplicated as the issue instructs rather than dragging in local-branch.ts. The one substantive change is swapping the test classifier from isTestFile (local-branch.ts) to isTestPath (signals/test-evidence.ts) in the shimmed path — CI is green across the full engine test suite plus the three pre-existing local-scorer/adapter/mcp test files, so behavioral parity is verified in practice even though it isn't visible from the diff alone.

Nits — 4 non-blocking

• The isTestFile → isTestPath swap in the re-implemented function (packages/gittensory-engine/src/local-scorer.ts) is a real dependency change, not a pure move — worth a one-line note in the PR description confirming isTestFile in local-branch.ts is itself just an alias for isTestPath, since that's the crux of why parity holds.
• The large header comment block is duplicated near-verbatim between src/signals/local-scorer.ts's shim comment and packages/gittensory-engine/src/local-scorer.ts's file-level comment — fine for now but will drift if one is updated and not the other.
• packages/gittensory-engine/test/local-scorer.test.ts imports from ../dist/index.js, so the new test is only meaningful after a build step; make sure the workspace test script (already validated per the PR) is the one CI actually runs, not a bare vitest invocation that would silently skip a stale dist.
• Consider a short comment in the 3 duplicated type definitions pointing back to local-branch.ts's canonical versions so a future contributor knows to keep them in sync (or add a lightweight structural-equality test between the two, similar to the engine-parity test already run for this PR).

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewer
Linked issue	✅ Linked	#4253
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: 13 registered-repo PR(s), 6 merged, 0 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	reyanthony062001-ops; Gittensor profile; 13 PR(s), 0 issue(s).
Gate result	✅ Passing	No configured blocker found.

Linked issue satisfaction

Addressed
The PR extracts computeLocalScorerTokens verbatim into packages/gittensory-engine/src/local-scorer.ts, narrowly duplicates the 3 dependent type shapes from local-branch.ts (rather than moving that file), turns src/signals/local-scorer.ts into a thin re-export shim, and exports the function/types from the engine package's index.ts — matching every deliverable in the issue and the referenced extract

Review context

• Author: reyanthony062001-ops
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: HTML, Python
• Official Gittensor activity: 13 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 approves — the gate is satisfied and CI is green.