feat(mcp): register loopover_check_improvement_potential stdio tool

[andriypolanski]

Summary
Register loopover_check_improvement_potential as a local stdio MCP tool in packages/loopover-mcp/bin/loopover-mcp.ts.
Proxy POST /v1/lint/improvement-potential (same endpoint the existing improvement-potential CLI already uses).
In-process InMemoryTransport coverage so Codecov instruments the .ts bin.
Tool count pin 89 → 90.
Closes #7759

Test plan
mcp-cli-improvement-potential-stdio.test.ts (register + POST proxy)
Tool count pin 89 → 90

[superagent-security]

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

[loopover-orb]

Tip

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-22 00:06:24 UTC}

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

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
This PR registers loopover_check_improvement_potential as a stdio MCP tool that proxies POST /v1/lint/improvement-potential, mirroring the existing checkSlopRiskShape pattern and the CLI's improvementPotentialCli (visible in loopover-mcp.js reference) which already hits the same endpoint. The zod shape, descriptor entry, and registerStdioTool wiring follow the established conventions for sibling proxy tools (e.g. loopover_check_slop_risk), and the test uses the same InMemoryTransport in-process pattern as other stdio tool tests. The tool-count pin bump (92→93) is consistent across both assertion sites in mcp-tool-rename-aliases.test.ts.

Nits — 6 non-blocking

• The external brief flags test/unit/mcp-cli-improvement-potential-stdio.test.ts:32 as a potential secret, but it's clearly the literal placeholder string "in-process-token" used by sibling tests too — false positive, not worth acting on.
• packages/loopover-mcp/bin/loopover-mcp.ts:1911 uses `input: any` for the handler parameter, consistent with every other registerStdioTool call in this file, so it's an existing pattern rather than a new issue, but still worth typing eventually.
• The magic numbers (400/2000 max lengths) mirror checkSlopRiskShape's existing bounds exactly, so flagging them as new/unexplained is misleading — they're copied conventions, not novel choices.
• Consider whether complexityDeltas/duplicationDeltas need any documentation in the tool description beyond what's already there, since they're new optional fields not present in the sibling slop-risk shape.
• No changes needed to the test file structure — it correctly follows the loopover_plan_repo_issues has no REST route, CLI command, or local stdio MCP tool #7764 entrypoint-guard pattern used by sibling stdio tool tests.
• Possible secret-shaped assignment in the diff (generic_secret_assignment) — Verify the value is not a real credential.

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	#7759
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: 184 registered-repo PR(s), 118 merged, 22 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	andriypolanski; Gittensor profile; 184 PR(s), 22 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Linked issue satisfaction

Addressed
The PR adds a registerStdioTool block for loopover_check_improvement_potential following the sibling pattern, uses stdioToolDescription for the description, proxies the existing POST /v1/lint/improvement-potential endpoint without duplicating HTTP logic, and includes a matching in-process test plus updated tool count assertions.

Review context

• Author: andriypolanski
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: Python, TypeScript, JavaScript, Rust, Cuda, Kotlin, MDX, Scala
• Official Gittensor activity: 184 PR(s), 22 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 approves — the gate is satisfied and CI is green.