feat(mcp): expose skipped-PR audit trail as a maintainer MCP tool

[nghetienhiep]

Summary

• Add loopover_get_skipped_pr_audit, a maintainer-authenticated, read-only MCP tool that surfaces the same skipped-PR audit trail already exposed by GET /v1/app/skipped-pr-audit (listPrVisibilitySkipAuditEvents), so an MCP client can triage "did the bot look at PR #N, and if not, why" without the dashboard.
• Extracted the shared reason enum (PUBLIC_SURFACE_SKIP_REASONS) and the per-reason remediation text (skippedPrAuditRemediation) out of src/api/routes.ts into src/signals/settings-preview.ts (next to PublicSurfaceSkipReason) so the HTTP route and the new MCP tool consume one canonical copy instead of two independently-maintained literal lists.
• Added the tool to the maintainer-triage MCP agent profile's recommendedTools.

Scope

• The PR title follows type(scope): short summary Conventional Commit format, for example fix(api): restore profile access checks.
• This PR is focused and does not mix unrelated backend, UI, MCP, docs, dependency, and deploy changes.
• This follows CONTRIBUTING.md and does not reintroduce GitHub Pages, VitePress, site/, or CNAME.
• I linked a currently open issue this PR resolves (e.g. Closes #123) — a linked open issue is required for every contributor PR.

Closes #5825

Validation

• git diff --check
• npm run actionlint
• npm run typecheck
• npm run test:coverage locally; codecov/patch requires ≥99% coverage of the lines AND branches you changed (aim for 100% on your diff so CI variance does not fail near the threshold). Global coverage is a non-blocking trend with a loose 90% backstop, not the gate.
• npm run test:workers
• npm run build:mcp
• npm run test:mcp-pack
• npm run ui:openapi:check
• npm run ui:lint
• npm run ui:typecheck
• npm run ui:build
• npm audit --audit-level=moderate
• New or changed behavior has unit/integration tests for new branches, fallback paths, and sanitizer boundaries

If any required check was skipped, explain why:

• ui:lint / ui:typecheck / ui:build were skipped: this PR touches no files under apps/loopover-ui/** — the only production files changed are src/api/routes.ts, src/mcp/server.ts, src/signals/settings-preview.ts, and packages/loopover-mcp/bin/loopover-mcp.js. npm run ui:openapi:check passed with no drift, confirming the (unchanged) /v1/app/skipped-pr-audit request/response shape still matches the generated OpenAPI spec.
• Also ran and confirmed green: npm run db:migrations:check, npm run db:schema-drift:check, npm run selfhost:env-reference:check, npm run miner:env-reference:check, npm run cf-typegen:check, npm run manifest:drift-check, npm run command-reference:check, npm run docs:drift-check, npm run engine-parity:drift-check, npm run test:engine-parity, npm run test:live-gate-parity, npm run test:driver-parity (no DB/env/wrangler-binding/docs changes in this PR, so all no-op green as expected).
• npm run test:coverage full run in my local sandbox reports 15 failing tests across 3 files (selfhost-ams-reporting.test.ts, miner-discover-cli.test.ts, miner-live-issue-snapshot.test.ts), all pre-existing and unrelated to this diff: the sqlite3 CLI binary is not installed in this sandbox (those tests shell out to it), and a real GITHUB_TOKEN is present in this shell's env and leaks into tests asserting no-token behavior. Confirmed identical failures on a clean stash of this branch at the same base commit before any of my changes. New coverage confirmed via test/unit/mcp-skipped-pr-audit.test.ts (11 cases covering: default no-filter feed scoped to the caller's own repos, each of repoFullName/reason/since filtered independently, limit clamped at both the 1 and 100 boundaries, an empty-page result, a session with no maintainer/owner/operator role being forbidden, a non-operator owner explicitly requesting its own scoped repo, a maintainer forbidden from a repo outside its scope, an unparseable since value, and the static mcp identity both forbidden and allowed depending on MCP_READ_REPO_ALLOWLIST) plus the existing test/unit/mcp-output-schemas.test.ts and test/integration/api.test.ts skipped-pr-audit route coverage, all still green after the shared-constant extraction. Verified via lcov that every new line and branch in src/mcp/server.ts (input/output schema, tool registration, requireSkippedPrAuditAccess, getSkippedPrAudit) and the new/moved code in src/signals/settings-preview.ts is hit.

Safety

• No secrets, wallet details, hotkeys, coldkeys, user PATs, private keys, raw trust scores, private rankings, or private maintainer evidence are exposed.
• Public GitHub text stays sanitized, low-noise, and does not imply compensation guarantees or optimization tactics.
• Auth, cookie, CORS, GitHub App, Cloudflare, or session changes include negative-path tests.
• API/OpenAPI/MCP behavior is updated and tested where needed.
• UI changes use live API data or real empty/error/loading states, not production mock/demo fallbacks.
• Visible UI changes include a UI Evidence section below with JPG/JPEG or PNG screenshots arranged as organized, captioned, clickable thumbnails. SVG screenshots are not used as review evidence. Review-only screenshots or recordings are not committed to the repository.
• Public docs/changelogs are updated where needed; changelogs are only edited for release-prep PRs.

No UI changes in this PR — the last two Safety checkboxes and the UI Evidence section below are not applicable.

UI Evidence

Not applicable — this PR only adds a backend/MCP tool with no visible UI, frontend, docs, or extension surface.

Notes

• The new tool's repo-scope gate (requireSkippedPrAuditAccess) mirrors the HTTP route's skippedPrAuditRepoScope (same loadControlPanelRoleSummary/loadControlPanelAccessScope calls, same maintainer/owner/operator role gate, same "no filter → caller's own scoped repos" fallback), reimplemented against this file's MCP identity/throw conventions rather than reused directly, since skippedPrAuditRepoScope is bound to a Hono ProtectedRouteContext and returns a Response, neither of which an MCP tool method can use. The shared static mcp CLI token additionally requires the unscoped MCP_READ_REPO_ALLOWLIST wildcard opt-in for this tool, matching every other cross-repo/maintainer-report MCP gate in this file (requireOperatorAccess, requireDiscoveryAccess).

[superagent-security]

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

[loopover-orb]

Warning

LoopOver found maintainer review notes

Scoped related-work signals were found for this PR. They are advisory unless the gate reports a blocker.

Readiness score: 57/100

Signal	Result	Evidence	Action
Linked issue	✅ Linked	#5825	No action.
Related work	⚠️ 3 scoped overlaps	Top overlaps are listed below; lower-confidence bulk is hidden.	Review top overlaps.
Change scope	❌ 8/20	High review scope from cached public metadata (1 linked issue).	Add a concise scope and risk note.
Validation posture	❌ 5/25	Preflight is holding this PR: the review lane is unavailable, so it is not ready for automated review.	Await review-lane availability.
Contributor workload	✅ 10/10	Author activity: 52 registered-repo PR(s), 23 merged, 0 issue(s).	No action.
Contributor context	✅ Confirmed Gittensor contributor	nghetienhiep; Gittensor profile; 52 PR(s), 0 issue(s).	No action.
Gate result	⚠️ Advisory only	Advisory only.	No action.

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.

Review context

• Author: nghetienhiep
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository registration is not available in the local LoopOver cache.
• Public profile languages: not available
• Official Gittensor activity: 52 PR(s), 0 issue(s).
• Related work: Titles/paths share 6 meaningful terms. (PR #5906)
• Related work: Titles/paths share 6 meaningful terms. (PR #5909)
• Related work: Titles/paths share 8 meaningful terms. (PR #5906, PR #5909)

Maintainer notes

• Repo lane is not ready for a confident recommendation: Repository registration is not available in the local LoopOver cache.
• Possible duplicate or overlapping work: 3 related open work cluster(s) were detected.

Contributor next steps

• Start here: Review top overlaps.
• Then work through the remaining 5 steps in the Signals table above.

• Re-run LoopOver review

---

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

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 95.16%. Comparing base (1b62bf0) to head (6e059b0).
⚠️ Report is 7 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #5905   +/-   ##
=======================================
  Coverage   95.15%   95.16%           
=======================================
  Files         589      589           
  Lines       46695    46723   +28     
  Branches    14902    14913   +11     
=======================================
+ Hits        44434    44462   +28     
  Misses       1511     1511           
  Partials      750      750           

Flag	Coverage Δ
shard-1	43.93% <10.52%> (-0.02%)	⬇️
shard-2	36.14% <86.84%> (+0.10%)	⬆️
shard-3	32.04% <10.52%> (-0.02%)	⬇️
shard-4	32.83% <10.52%> (-0.04%)	⬇️
shard-5	31.49% <10.52%> (-0.02%)	⬇️
shard-6	44.53% <34.21%> (-0.03%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Files with missing lines	Coverage Δ
src/api/routes.ts	94.40% <ø> (-0.03%)	⬇️
src/mcp/server.ts	96.06% <100.00%> (+0.12%)	⬆️
src/signals/settings-preview.ts	98.73% <100.00%> (+0.08%)	⬆️