feat(mcp): add remote + stdio + CLI for selftune override audit

[RealDiligent]

Summary

• Register loopover_get_selftune_override_audit as a remote MCP tool (maintainer category) calling listOverrideAudit, with optional limit.
• Register the matching local stdio MCP tool proxying GET .../selftune/overrides/audit.
• Add maintain selftune-audit [--limit N] CLI verb mirroring the precision emit/--json shape.
• Update PowerShell completer pin for the new maintain subcommand.
• Rebased onto main after loopover_get_repo_focus_manifest (loopover_get_repo_focus_manifest (read-only) has a REST route but no remote MCP tool or local stdio MCP tool #7808/feat(mcp): add remote + stdio for loopover_get_repo_focus_manifest (#7808) #7943); tool count pin 89 → 90.

Closes #7798

Test plan

• mcp-selftune-override-audit.test.ts (authorized + empty + allowlist deny)
• mcp-cli-selftune-override-audit.test.ts (in-process stdio, with/without limit)
• mcp-cli-maintain.test.ts selftune-audit plain/json/limit
• mcp-cli-basics.test.ts PowerShell completer pin
• Tool count pin 89 → 90

[superagent-security]

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

[codecov]

Codecov Report

❌ Patch coverage is 59.09091% with 9 lines in your changes missing coverage. Please review.
✅ Project coverage is 82.70%. Comparing base (bc042ed) to head (9d8bc0f).
⚠️ Report is 2 commits behind head on main.

Files with missing lines	Patch %	Lines
packages/loopover-mcp/bin/loopover-mcp.ts	35.71%	8 Missing and 1 partial ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #7947      +/-   ##
==========================================
+ Coverage   79.50%   82.70%   +3.20%     
==========================================
  Files         735       92     -643     
  Lines       75452    24776   -50676     
  Branches    22952     4738   -18214     
==========================================
- Hits        59986    20492   -39494     
+ Misses      12710     4047    -8663     
+ Partials     2756      237    -2519     

Flag	Coverage Δ
shard-1	22.89% <36.36%> (-31.34%)	⬇️
shard-2	22.40% <45.45%> (-31.39%)	⬇️
shard-3	33.61% <45.45%> (?)

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

Files with missing lines	Coverage Δ
src/mcp/server.ts	96.73% <100.00%> (+9.69%)	⬆️
packages/loopover-mcp/bin/loopover-mcp.ts	14.32% <35.71%> (+0.22%)	⬆️

... and 643 files with indirect coverage changes

[loopover-orb]

Caution

🛑 LoopOver review result - fixes required

_{Review updated: 2026-07-21 21:00:32 UTC}

10 files · 1 AI reviewer · no blockers · CI failing · unstable

🛑 Suggested Action - Fix Blockers

Review summary
This PR mirrors the established pattern (like gate-precision) to add a new maintainer-only tool surfacing the self-tune override audit trail across remote MCP, stdio MCP, and CLI, with pin updates for tool count (89→90) and the PowerShell completer. The wiring traces correctly: the remote tool calls listOverrideAudit from src/review/auto-apply.ts behind requireRepoAccess, the stdio tool proxies GET .../selftune/overrides/audit, and the CLI subcommand mirrors precision's emit/--json shape. Test coverage is broad (authorized/empty/allowlist-deny, stdio in-process, CLI plain/json/limit), though the CLI's `Number(options.limit)` producing `NaN > 0 === false` for an omitted --limit is untested but behaves correctly by falling through.

Nits — 6 non-blocking

• packages/loopover-mcp/bin/loopover-mcp.ts:2833 and :3608 use `any` for the callback/handler parameter, consistent with the rest of the file's style but still worth tightening if this file is ever split.
• The flagged 'secret' in test/unit/mcp-cli-selftune-override-audit.test.ts:30 is just `LOOPOVER_API_TOKEN = "in-process-token"`, a hardcoded test fixture value, not a real leaked credential.
• codecov/patch is failing (45–59% vs 99% target); the two full-file-content dumps show the changed lines are exercised by the new tests, so this is likely coverage-tool line-counting on the diff context rather than truly uncovered branches — worth confirming which specific lines are flagged.
• The `// Self-tune override audit trail (selftune/overrides/audit) has no remote MCP tool, local stdio MCP tool, or CLI verb #7798` comment style at packages/loopover-mcp/bin/loopover-mcp.ts:2822-2823 duplicates explanation already present in the server.ts comment; fine but slightly repetitive across the two proxy layers.
• Double check codecov/patch's uncovered-line report directly rather than assuming it's noise — if it names lines in src/mcp/server.ts's getSelftuneOverrideAudit or the stdio proxy, add a targeted test.
• Possible secret-shaped assignment in the diff (generic_secret_assignment) — Verify the value is not a real credential.

CI checks failing

• codecov/patch — 59.09% of diff hit (target 99.00%)

Decision drivers

• ✅ Code review — No blockers (1 reviewer)
• ✅ Gate result — Passing (No configured blocker found.)

Context & advisory signals — never blocks the verdict

Signal	Result	Evidence
Linked issue	✅ Linked	#7798
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: 395 registered-repo PR(s), 173 merged, 36 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	RealDiligent; Gittensor profile; 395 PR(s), 36 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Linked issue satisfaction

Addressed
The PR registers loopover_get_selftune_override_audit as a remote MCP tool (maintainer category) calling listOverrideAudit, adds the matching stdio tool proxying GET .../selftune/overrides/audit, and adds a maintain selftune-audit CLI verb mirroring precision's emit/--json structure, exactly the three mirror surfaces requested.

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: Python, JavaScript, Ruby, Svelte, TypeScript, Markdown, MDX
• Official Gittensor activity: 395 PR(s), 36 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.

🧪 Chat with LoopOver

Ask LoopOver a question about this PR directly in a comment — grounded only in the same cached, public-safe facts shown above, never a new claim.

• @loopover ask <question> answers contribution-quality Q&A with source citations and freshness.
• @loopover chat <question> answers in natural prose from cached decision-pack facts via local inference (maintainer/collaborator; read-only).
• A plain-language @loopover mention with a real question is routed to the closest matching read-only command automatically — no exact syntax required.

Full command reference: https://loopover.ai/docs/loopover-commands

🧪 Experimental — new and may change.

_{🟩 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 LoopOver, a quiet PR intelligence layer for OSS maintainers.

• Re-run LoopOver review

[loopover-orb]

LoopOver is closing this pull request on the maintainer's behalf (CI is failing (codecov/patch)). 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.