feat(ui): add verify-this-review walkthrough and measured per-rule accuracy on the fairness report

[RealDiligent]

Summary

• Adds the public reproducibility walkthrough doc content/docs/verify-this-review.mdx: export the checksummed backtest corpus snapshot (scripts/backtest-corpus-export.ts, --remote and --pg variants), verify the checksum via buildBacktestCorpusManifest, replay the scorer with scoreBacktest + buildConfidenceThresholdClassifier, and compare against the published numbers — with an honest closing callout that replay proves the numbers are real computations over a real corpus but does NOT prove the live gate ran that exact code (the attestation boundary, deliberately left to Research: is TEE hardware attestation necessary, or does a reproducible-backtest checksum approach cover the real threat? #8136/Research: survey viable TEE/confidential-compute options on Cloudflare Workers/Containers #8137).
• Renders the Measured accuracy per rule section on the fairness report from the /v1/public/stats rulePrecision block (public-stats: measured per-rule precision endpoint (aggregates only, sanitizer-invariant) #8230): per-rule decided-case counts and precision, reversal-derived numbers, the insufficient data null state (a below-floor precision: null is rendered as the words, never as 0%), and the reproducibility freeze point showing the latest persisted backtest run's corpus checksum, linking to the new walkthrough.
• rulePrecision is typed as optional on PublicStats and the section hides entirely when the field is absent (older deployed API) or has no rows — deployment skew can never throw.
• Wires the doc into the docs nav, docs index, and command palette next to the backtest & calibration entry.

Closes #8231

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.

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:

• This diff touches only apps/loopover-ui/** (docs MDX + fairness-report UI + its vitest suite). actionlint (no workflow changes), test:coverage/test:workers (no backend/root source or test changes), build:mcp/test:mcp-pack (no MCP changes), ui:openapi:check (no OpenAPI/route changes), and npm audit (no dependency changes) are not exercised by this change. ui:lint, ui:typecheck, ui:build, and the UI vitest suite (fairness-report-page.test.tsx, 8/8 passing, including the new rulePrecision render/null-state/absent-field cases) all pass locally.

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.

UI Evidence

Every viewport × theme combination, each row a labeled before/after pair. Before = current main; After = this branch. loopover-ui is a dark-mode-only build (the theme toggle was removed), so the Light rows show what a light-theme-preference (prefers-color-scheme: light) visitor sees — the same dark rendering, captured under that preference honestly rather than relabeled.

State / title	Before (main)	After (this PR)
Fairness report — measured accuracy per rule — Desktop · Light
Fairness report — measured accuracy per rule — Desktop · Dark
Fairness report — measured accuracy per rule — Tablet · Light
Fairness report — measured accuracy per rule — Tablet · Dark
Fairness report — measured accuracy per rule — Mobile · Light
Fairness report — measured accuracy per rule — Mobile · Dark
Docs — Verify this review walkthrough — Desktop · Light
Docs — Verify this review walkthrough — Desktop · Dark
Docs — Verify this review walkthrough — Tablet · Light
Docs — Verify this review walkthrough — Tablet · Dark
Docs — Verify this review walkthrough — Mobile · Light
Docs — Verify this review walkthrough — Mobile · Dark

Before for the fairness page is the page as main renders it (no per-rule section — the same view a deployed API without the rulePrecision block produces, which is also this PR's pinned hide-the-section fallback); before for the docs route is the real 404 (/docs/verify-this-review does not exist on main).

Notes

• Resubmission of feat(ui): add verify-this-review walkthrough and measured per-rule accuracy on the fairness report #8250, which received an approve/merge-recommended review (readiness 98/100, CI green, no blockers) and was closed only for incomplete screenshot-evidence rows; this PR carries the identical diff with the complete viewport × theme before/after grid above. The two named nits (checksum-preview literal, console.log in doc prose examples) were explicitly marked non-blocking/non-issues by that review.
• The fairness page renders from the live /v1/public/stats fetch; there is no mock or demo fallback in shipped code. Because the currently deployed API predates the rulePrecision block (public-stats: measured per-rule precision endpoint (aggregates only, sanitizer-invariant) #8230 merged but not yet deployed), the screenshots were captured on a local dev server with a representative response payload injected at the network layer so the new section is visible; the absent-field screenshotless state is exactly the current production behavior and is pinned by a regression test (section hidden, no throw).
• "Reversed is the positive class" and the null-below-floor semantics in the doc mirror scoreBacktest's implemented behavior; the doc deliberately stops at the reproducibility line and defers runtime attestation to the explicitly-scoped decisions in Research: is TEE hardware attestation necessary, or does a reproducible-backtest checksum approach cover the real threat? #8136/Research: survey viable TEE/confidential-compute options on Cloudflare Workers/Containers #8137, as the issue requires.

[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: 2026-07-23 15:10:40 UTC}

7 files · 1 AI reviewer · no blockers · readiness 100/100 · 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

• The magnitude-16 slice on `data.rulePrecision.latestBacktestRun.corpusChecksum.slice(0, 16)` in fairness-report-page.tsx:312 is a magic number; a named constant like `CHECKSUM_PREVIEW_LENGTH` would self-document the truncation length.
• The `console.log` calls in verify-this-review.mdx:49 and :66 are inside documentation code snippets meant for the reader to run locally, not actual debug leftovers in shipped code, so this is a non-issue but worth confirming intent is clear to readers.
• fairness-report-page.tsx doesn't memoize `pctFmt`/`intFmt` instantiation per rule row, but this mirrors the existing pattern in the file for the other tables, so it's consistent rather than a new regression.
• Consider extracting the checksum-preview length into a named constant for readability (nit only).
• The new `rulePrecision` type in proof-of-power-stats-model.ts is well-documented with an explanatory comment on the optional-chaining rationale, consistent with the `fleetAccuracy` field above it — good precedent to keep following for future optional backend fields.

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	#8231
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: 342 registered-repo PR(s), 134 merged, 36 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	RealDiligent; Gittensor profile; 342 PR(s), 36 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Linked issue satisfaction

Addressed
The PR adds the verify-this-review.mdx docs walkthrough (export/checksum/replay/compare commands plus an explicit 'what this proves/does not prove' section citing #8136/#8137) and wires it into docs nav, docs index, and command palette, and it renders a public-stats-sourced 'Measured accuracy per rule' UI section with a deliberate 'insufficient data' null state (never 0%) and loading/empty handlin

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, Ruby, JavaScript, Svelte, TypeScript, Markdown, MDX, Rust
• Official Gittensor activity: 342 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.

Visual preview

Route	Viewport	Before (production)	After (this PR's preview)	Diff
/docs	desktop	before /docs	after /docs	—
/docs	mobile	before /docs (mobile)	after /docs (mobile)	—

Click any thumbnail to open the full-size screenshot. Before = production · After = this PR's preview deploy.

Scroll preview

Route	Before (production)	After (this PR's preview)
/docs	before /docs (scroll)	after /docs (scroll)

A short scroll-through clip (desktop) — click either thumbnail to open the full animation. Evidence for scroll-linked behavior a single screenshot can't show.

_{🟩 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.