feat(mcp): stream loop progress to the customer via a progress snapshot

[luciferlive112116]

Summary

Implements #4800 — a customer can watch their rented loop's progress in near-real time instead of polling for a final result. Adds the deterministic brain of the stream (part of the Rent-a-Loop path #4778), so a customer surface (#4807) pushes on change rather than on a fixed interval.

• new packages/loopover-engine/src/loop-progress.ts (pure): buildProgressSnapshot(state) formats a running loop's already-computed state into phase, status, iteration + percentComplete (through the iteration budget — capped at 100, null when the budget is unknown), and a bounded recent-activity tail. progressChanged(prev, next) decides when a snapshot has meaningfully changed on the displayed axes — so the surface streams on change, not on a fixed interval (the exact acceptance criterion). No IO, no transport.
• new loopover_build_progress_snapshot MCP tool (src/mcp/server.ts); src/loop-progress.ts is a thin re-export shim over the engine module.

Acceptance (#4800): a test loop's progress is visible and updating without the client polling on a fixed interval — the progressChanged push-on-change model provides exactly that, tested at the engine level and end-to-end through the MCP tool.

Closes #4800

Scope

• Conventional Commit title (feat(mcp): …).
• Focused: the progress-snapshot model + its tool, on the merged intake/claim/results chain.
• Follows CONTRIBUTING.md; no site//CNAME/VitePress.
• Linked open issue: Closes #4800.

Validation

• git diff --check
• npm run typecheck clean; npm --workspace @loopover/engine run build + npm run build:mcp clean
• npm run test:coverage on the changed code: loop-progress.ts 100% lines & branches (14/14); the new src/mcp/server.ts handler + schema + registration fully covered (all 43 changed lines, both branch sides — diff-verified).
• MCP tool-invariant suites pass with the new tool (mcp-output-schemas "outputSchema on EVERY tool" + "schema-valid structured content").
• Tests cover known/unknown/zero/over-budget percent-complete, the activity-tail cap, the done transition, and every progressChanged axis (first snapshot, phase/status/iteration/activity change, no-change).

If any required check was skipped, explain why:

• Full test:ci not run end-to-end locally (Linux-only shell/self-host steps on Windows); the change-relevant gates above were validated directly.

Safety

• No secrets, wallet/hotkey/coldkey, trust scores, rewards, private rankings, or private maintainer evidence — the snapshot is public-safe loop status only, source-free.
• No auth/cookie/CORS/GitHub App/session change (pure function over caller-supplied data).
• MCP behavior added + tested (output schema + schema-valid content).
• No UI changes; no changelog edit.

[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 95.11%. Comparing base (9b3f4b2) to head (f30f682).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #5798   +/-   ##
=======================================
  Coverage   95.10%   95.11%           
=======================================
  Files         584      585    +1     
  Lines       46440    46455   +15     
  Branches    14849    14854    +5     
=======================================
+ Hits        44169    44184   +15     
  Misses       1515     1515           
  Partials      756      756           

Flag	Coverage Δ
shard-1	43.74% <26.66%> (-0.02%)	⬇️
shard-2	36.25% <26.66%> (+0.06%)	⬆️
shard-3	32.12% <73.33%> (-0.03%)	⬇️
shard-4	33.46% <26.66%> (+0.54%)	⬆️
shard-5	31.07% <80.00%> (-0.57%)	⬇️
shard-6	44.51% <26.66%> (-0.01%)	⬇️

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

Files with missing lines	Coverage Δ
packages/loopover-engine/src/loop-progress.ts	100.00% <100.00%> (ø)
src/mcp/server.ts	95.90% <100.00%> (+0.03%)	⬆️

🚀 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

🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩🟩

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-14 17:38:07 UTC}

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

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
This is a pure, small, well-tested addition: a `buildProgressSnapshot`/`progressChanged` module plus a thin MCP tool wrapper, closing issue #4800 with a push-on-change progress model. The math and edge cases (null/0/undefined maxIterations, capping at 100, activity-tail slicing) are correctly handled and covered by both engine-level and MCP end-to-end tests. The `src/loop-progress.ts` re-export shim mirrors the existing `results-payload.ts`/`idea-intake.ts` pattern, and CI is green across build/typecheck/tests.

Nits — 5 non-blocking

• packages/loopover-engine/src/loop-progress.ts:24-25 — recentActivity/detail/at use `| undefined` in addition to `?:` which is redundant under standard TS config; harmless but inconsistent with typical style elsewhere in the codebase.
• test/unit/loop-progress.test.ts — `progressChanged` only compares `recentActivity.length`, not content; a test that swaps activity items of the same length (e.g. replacing the last entry) would silently not push — worth a comment or explicit test acknowledging this is an intentional simplification.
• src/mcp/server.ts:987 — `recentActivity` cap of `.max(1000)` in the input schema is much looser than the internal `MAX_PROGRESS_ACTIVITY` (10) truncation; consider aligning or documenting why the input allows more than is ever kept.
• The 'size smells' flags on packages/loopover-engine/src/index.ts and src/mcp/server.ts reflect pre-existing file growth, not something this diff meaningfully worsens (it adds ~10 and ~43 lines respectively) — not actionable against this PR.
• Consider exporting `MAX_PROGRESS_ACTIVITY` from `src/loop-progress.ts` shim explicitly if any downstream CLI needs the same cap value referenced by name (currently only re-exported via `export *`, which is fine but worth double-checking is intended).

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewer
Linked issue	✅ Linked	#4800
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: 91 registered-repo PR(s), 37 merged, 39 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	luciferlive112116; Gittensor profile; 91 PR(s), 39 issue(s).
Gate result	✅ Passing	No configured blocker found.
Improvement	✅ Minor	risk: clean · value: minor — Code changes are accompanied by test evidence. LLM value judgment: moderate — The PR adds a focused, pure, fully-tested progress-snapshot model plus one MCP tool that directly closes the linked issue #4800 without scope creep into unrelated areas.

Linked issue satisfaction

Partially addressed
The PR adds a deterministic snapshot builder and change-detection logic plus a synchronous MCP tool to compute a snapshot on request, but this is a pull-based tool call, not an actual streaming/push mechanism — a client still must invoke the tool repeatedly to see updates, so the acceptance criterion 'without the client needing to poll on a fixed interval' is only partially met since polling on-de

Review context

• Author: luciferlive112116
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: Python, JavaScript, MDX, Rust, TypeScript
• Official Gittensor activity: 91 PR(s), 39 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.

[BETA] Chat with Gittensory

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

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

Full command reference: https://gittensory.aethereal.dev/docs/gittensory-commands

_{🟩 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]

Gittensory approves — the gate is satisfied and CI is green.