feat(review): capture bounded raw context in configured-blocker fired events, permanently excluding secret_leak

[cleanjunc]

Summary

Closes #8130.

Extends #8104's generic recordConfiguredGateBlockerSignals (src/rules/advisory.ts) so every non-excluded configured-blocker code captures the raw context its detection actually evaluated — making those corpora rich enough to backtest logic/detection changes, not just thresholds — while permanently excluding secret_leak by design.

• RAW_CONTEXT_EXCLUDED_CODES — the exported, named constant the issue mandates (secret_leak inside), with the full security reasoning attached: capturing the diff that triggered a secret_leak finding would store the leaked credential itself in the calibration audit trail. The comment also carries the issue's Boundaries note (additions require an explicit, maintainer-reviewed redaction design — never a quiet edit).
• RAW_CONTEXT_MAX_DIFF_CHARS = 120000 — mirrors src/services/ai-review.ts's own input.diff.slice(0, 120000) bound exactly (the same number, with a keep-in-sync comment), never a newly invented bound.
• AI-judgment codes (ai_consensus_defect/ai_review_split): the metadata gains the AI review's own diff, threaded from the one caller (src/queue/processors.ts) via the already-memoized getReviewFiles() + the same buildAiReviewDiff the review consumed — never re-fetched or re-derived.
• Every other non-excluded code — audited individually per the issue's requirement (the audit is recorded in the function's doc comment): missing_linked_issue reads linkage state, duplicate_pr_risk reads sibling-PR overlap, pre_merge_check_required/cla_check_unresolved read check-run conclusions, manifest_missing_tests reads changed paths vs manifest expectations, the review-thread code reads unresolved-thread state — none evaluates raw diff content, so each captures its detection's own recorded detail (the narrative of exactly that evaluated signal) as rawSignal, same bound.
• linked_issue_scope_mismatch remains skipped entirely (calibration: wire linked_issue_scope_mismatch into the shared signal-tracking module (fired + reversal) #8101's own site records it — unchanged), and the calibration: wire recordRuleFired/recordHumanOverride for the remaining isConfiguredGateBlocker codes (excludes linked_issue_scope_mismatch, covered by #8101) #8104 behavior for confidence capture is preserved byte-for-byte.

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:

• Ran the directly-affected gates: test/unit/configured-gate-blocker-signals.test.ts — 14/14 green (8 pre-existing calibration: wire recordRuleFired/recordHumanOverride for the remaining isConfiguredGateBlocker codes (excludes linked_issue_scope_mismatch, covered by #8101) #8104 cases unchanged + 6 new calibration: capture bounded raw context for the remaining isConfiguredGateBlocker codes, excluding secret_leak #8130 cases: the security negative asserting secret_leak's event carries NO diff/rawSignal even with confidence, detail, and a threaded diff all present; the AI-code diff capture with the truncation branch at the exact bound; the no-diff-threaded branch; the audited rawSignal fallback; the empty-detail/no-metadata branch; the linked_issue_scope_mismatch skip) — plus npm run engine-parity:drift-check (ok: the advisory.ts↔gate-advisory.ts twin pair still agrees; the added recorder context is host-side-only, the same posture calibration: wire recordRuleFired/recordHumanOverride for the remaining isConfiguredGateBlocker codes (excludes linked_issue_scope_mismatch, covered by #8101) #8104 itself established). Real CI re-runs typecheck and the sharded suite here.
• workers/mcp/ui/audit/actionlint: untouched surfaces — two source files + one test file; zero dependency changes.

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.

The three unchecked Safety boxes are N/A: no auth/session/UI change. Note the change is itself security-motivated in part: the secret_leak exclusion exists precisely so no leaked credential can ever enter the calibration audit trail, and that property is pinned by a dedicated negative test.

UI Evidence

N/A — no UI change.

Notes

• The metadata builder intentionally emits NO metadata key at all (rather than an empty object) when nothing is captured — preserving calibration: wire recordRuleFired/recordHumanOverride for the remaining isConfiguredGateBlocker codes (excludes linked_issue_scope_mismatch, covered by #8101) #8104's exact optional-property behavior for that case.

[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 90.37%. Comparing base (c02a277) to head (0ef6ac5).
⚠️ Report is 6 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8135      +/-   ##
==========================================
- Coverage   92.02%   90.37%   -1.66%     
==========================================
  Files         760       99     -661     
  Lines       77394    26010   -51384     
  Branches    23392     5108   -18284     
==========================================
- Hits        71225    23506   -47719     
+ Misses       5061     2233    -2828     
+ Partials     1108      271     -837     

Flag	Coverage Δ
shard-1	79.48% <83.33%> (+22.38%)	⬆️
shard-2	34.97% <50.00%> (-19.37%)	⬇️
shard-3	40.82% <100.00%> (-9.51%)	⬇️

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

Files with missing lines	Coverage Δ
src/queue/processors.ts	95.76% <100.00%> (ø)
src/rules/advisory.ts	97.95% <100.00%> (+0.07%)	⬆️

... and 661 files with indirect coverage changes

[loopover-orb]

Tip

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-23 00:30:52 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 extends recordConfiguredGateBlockerSignals to capture bounded raw-context metadata (the AI review's diff for ai_consensus_defect/ai_review_split, or the finding's own detail string for every other non-excluded code) while explicitly hard-excluding secret_leak via a named, exported RAW_CONTEXT_EXCLUDED_CODES set. The exclusion logic in src/rules/advisory.ts:1108-1115 is correctly ordered (confidence is set first, then the excluded-code check gates diff/rawSignal capture), and the new caller wiring in src/queue/processors.ts threads buildAiReviewDiff(await getReviewFiles()) through the already-memoized file fetch rather than re-fetching. Test coverage in configured-gate-blocker-signals.test.ts directly exercises the security-critical secret_leak exclusion path plus both AI-code and non-AI-code capture branches.

Nits — 5 non-blocking

• src/rules/advisory.ts:1108 reuses RAW_CONTEXT_MAX_DIFF_CHARS (120000) to bound `finding.detail` (rawSignal) even though that constant's stated purpose is mirroring the AI review's diff-slice bound — consider a separate, smaller constant for the detail-string bound since detail text is never diff-sized.
• src/queue/processors.ts: buildAiReviewDiff(await getReviewFiles()) is now called on every `if (evaluation)` pass regardless of whether any AI-judgment finding is present in configuredBlockers — relies entirely on getReviewFiles() being memoized to stay cheap; worth a one-line comment noting that dependency so a future refactor doesn't silently drop the memoization.
• The doc comment on recordConfiguredGateBlockerSignals says the AI diff is 'never re-fetched or re-derived,' but the call site actually re-derives it via buildAiReviewDiff(await getReviewFiles()) rather than reusing a stored diff variable from the earlier AI review step — technically correct (same memoized files, same pure function) but the wording overstates it slightly.
• Consider renaming RAW_CONTEXT_MAX_DIFF_CHARS to something bound-agnostic (e.g. RAW_CONTEXT_MAX_CHARS) since it now bounds both diff and non-diff (detail) content.
• A short comment at the buildAiReviewDiff(await getReviewFiles()) call site confirming getReviewFiles() is memoized would help future readers avoid assuming this is a fresh fetch per gate evaluation.

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	#8130
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: 51 registered-repo PR(s), 16 merged, 33 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	cleanjunc; Gittensor profile; 51 PR(s), 33 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Linked issue satisfaction

Addressed
The diff adds an exported RAW_CONTEXT_EXCLUDED_CODES constant with secret_leak and the required rationale comment, threads the bounded AI review diff for ai_consensus_defect/ai_review_split from the existing caller without re-fetching, captures a detail-based rawSignal fallback for non-diff codes after an audit noted in the doc comment, and includes tests covering the secret_leak negative case, th

Review context

• Author: cleanjunc
• 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: 51 PR(s), 33 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.