feat(miner-selfimprove): read-only calibration dashboard view

[jeffrey701]

Closes #4261

What

computePhase7CalibrationLoop (phase7-calibration-loop.ts) already produces the tracked calibration metric — predicted-gate accuracy combined from the historical_replay and pr_outcome signals, scored against the documented 0.62 baseline. This adds the read-only view that projects a Phase7CalibrationLoopResult into a human-readable dashboard shape (a CLI table, or a panel in the miner dashboard, renders the rows).

Pure — no new computation. It re-shapes an already-computed result (predicted-gate accuracy vs realized pr_outcome is presented, never recomputed here — that stays in the calibration loop). Public-safe: only accuracies, sample sizes, freshness, and the loop's holdReasons are surfaced (no raw scores/rewards).

API (packages/gittensory-engine/src/calibration-dashboard.ts)

• resolveCalibrationDashboardStatus(result) → "on_track" | "below_baseline" | "insufficient_signal" | "disabled".
• buildCalibrationDashboardView(result) → { status, headline, rows, holdReasons } — a headline (e.g. "68% combined (+6pts vs baseline)") plus labeled rows: combined accuracy, signed delta-from-baseline, each source (accuracy + n= sample size + fresh/stale), the replay-harness state, and whether an autonomy increase is permitted. holdReasons pass through verbatim so an operator sees why an increase is (or isn't) allowed.

Files

• packages/gittensory-engine/src/calibration-dashboard.ts — the view.
• packages/gittensory-engine/src/index.ts — barrel re-export.
• test/unit/calibration-dashboard.test.ts — tests.

Testing

npx vitest run test/unit/calibration-dashboard.test.ts
npm run typecheck

5/5 tests pass (on-track, below-baseline negative-delta, insufficient-signal em-dash, and disabled/harness-hold/replay-due branches); typecheck clean. New engine file at 100% line + branch coverage (12/12, 24/24, 5/5).

[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 94.03%. Comparing base (23b3549) to head (cfd8e16).
⚠️ Report is 3 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #4504   +/-   ##
=======================================
  Coverage   94.03%   94.03%           
=======================================
  Files         420      421    +1     
  Lines       37524    37541   +17     
  Branches    13702    13714   +12     
=======================================
+ Hits        35285    35302   +17     
  Misses       1583     1583           
  Partials      656      656           

Files with missing lines	Coverage Δ
...ges/gittensory-engine/src/calibration-dashboard.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 22:20:52 UTC}

3 files · 1 AI reviewer · no blockers · readiness 82/100 · CI green · clean

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
The AI review returned non-blocking notes for this change but did not include a separate narrative summary. Review the nits below before deciding this PR.

Nits — 5 non-blocking

• calibration-dashboard.ts:64-90 always renders `Historical replay`/`PR outcome`/`Replay harness`/`Autonomy increase` rows even when `status` is `disabled` or `insufficient_signal`, which can show stale-looking `n=0`/`stale` detail alongside a 'disabled' headline — worth a one-line note in the row detail or a doc comment on intended consumer behavior in that case.
• test/unit/calibration-dashboard.test.ts imports from `../../packages/gittensory-engine/src/index` (no `.js`) rather than the `.js`-suffixed relative imports used inside the package itself — confirm this matches the existing test-import convention elsewhere in `test/unit/`.
• calibration-dashboard.ts:9-13 duplicates the module-level doc comment already given in the PR description almost verbatim; consider trimming to what a reader can't get from the exported JSDoc on the two functions.
• Consider whether `rows` should omit or clearly flag source rows when `status === "disabled"`, so a CLI/UI renderer doesn't need its own special-casing on top of `status`.
• If a future issue references linking `holdReasons` codes to human text, this view is the natural place to add a `label`/description lookup — not needed now given the pure read-only scope.

Signal	Result	Evidence
Code review	✅ No blockers	1 reviewer
Linked issue	✅ Linked	#4261
Related work	⚠️ 1 scoped overlap	Top overlaps are listed below; lower-confidence bulk is hidden.
Change scope	❌ 8/20	High 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: 106 registered-repo PR(s), 45 merged, 4 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	jeffrey701; Gittensor profile; 106 PR(s), 4 issue(s).
Gate result	✅ Passing	No configured blocker found.

Linked issue satisfaction

Partially addressed
The PR delivers the read-only dashboard view with the requested combined accuracy/baseline, per-source freshness, holdReasons, and replayRunDue rows, and includes tests for healthy, below-baseline, insufficient-signal, and disabled/hold states. However, the tests hand-construct `Phase7CalibrationLoopResult` objects directly rather than building `HistoricalReplayCalibrationInput`/`PrOutcomeCalibrat

Review context

• Author: jeffrey701
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: not available
• Official Gittensor activity: 106 PR(s), 4 issue(s).
• Related work: Titles/paths share 6 meaningful terms. (issue #4261, issue #4307)

Contributor next steps

• Review top overlaps.
• Add a concise scope and risk note.
• Check active issues and PRs before submitting.

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.