fix(miner): drop discovery-index candidates whose repo bans AI contributions (#9680)

[kai392]

What

supplementWithDiscoveryIndex folded hosted discovery-index candidates into the local fan-out with dedupe as the only filter. The index client deliberately preserves a repo's AI-contribution ban (normalizeDiscoveryIndexCandidate writes aiPolicyAllowed !== false), but nothing downstream re-checks it — and the as RawCandidateIssue cast (whose aiPolicyAllowed is declared as the literal true) laundered a false straight through. So an index-served candidate flagged "this repo bans automated contributions" was ranked and enqueued into the miner's own portfolio backlog. The local path enforces the ban hard (fetchTargetIssues: if (!verdict.allowed) return []); the index path did not.

How

• discover-cli.ts: filter out aiPolicyAllowed === false candidates before the dedupe filter / cast, mirroring fetchTargetIssues's own drop. The predicate is !== false, so a candidate that omits the field entirely is still kept.
• discovery-index-client.ts: recordDiscoveryTelemetry now accepts an optional droppedAiBanned count, spread conditionally into the log payload so every existing caller keeps its exact { event, outcome } shape; the discover_query call reports the number dropped.

Tests

• One index candidate aiPolicyAllowed: false + one true → only the allowed one reaches ranked/enqueued output, and droppedAiBanned: 1 is reported (fails against current code).
• A candidate that omits aiPolicyAllowed is still kept (not a fail-closed drop).
• A direct recordDiscoveryTelemetry case (both opt-ins enabled, count supplied) asserting the payload carries the count — so both arms of the conditional spread are covered.

The shared-module test that asserts the exact { event, outcome } telemetry payload still passes.

Closes #9680

[loopover-orb]

Tip

✅ LoopOver review result - approve/merge recommended

_{Review updated: 2026-07-29 21:46:38 UTC}

4 files · 1 AI reviewer · no blockers · readiness 98/100 · CI green · clean

✅ Suggested Action - Approve/Merge

• safe to merge

Review summary
This PR closes the fix in the right layer: it drops discovery-index candidates with aiPolicyAllowed === false before the dedupe/cast in supplementWithDiscoveryIndex, mirroring fetchTargetIssues' local enforcement, and threads a droppedAiBanned count through recordDiscoveryTelemetry via a conditional spread that preserves the exact prior payload shape for existing callers. The trace holds up: the `as RawCandidateIssue` cast (whose type declares aiPolicyAllowed as literal true) previously laundered a false straight through with no re-check downstream, so this is a real reachable defect, not a can't-occur fix. Tests cover both arms (banned dropped, field-omitted kept) and the telemetry spread's both branches, and the PR closes #9680 as required.

Nits — 4 non-blocking

• nit: discover-cli.ts's inline comment above the filter is quite long (6+ lines) for a single filter predicate — consider trimming to the core rationale.
• nit: the filter predicate `candidate.aiPolicyAllowed !== false` is duplicated conceptually from fetchTargetIssues' own ban check; consider extracting a small shared helper if this pattern recurs a third time.
• Consider adding a one-line JSDoc/type note on DiscoveryIndexCandidate's aiPolicyAllowed field itself (if not already documented) so future readers don't need to trace through discover-cli.ts to learn that omission means allowed.
• The droppedAiBanned count could also be useful in the non-JSON renderDiscoverSummary output for operator visibility, though this is out of scope for this fix.

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	#9680
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: 95 registered-repo PR(s), 50 merged, 7 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	kai392; Gittensor profile; 95 PR(s), 7 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: significant

Linked issue satisfaction

Addressed
The diff filters index candidates with aiPolicyAllowed === false before the dedupe/cast in supplementWithDiscoveryIndex, keeps field-absent candidates via the `!== false` predicate, and surfaces the dropped count via an extended recordDiscoveryTelemetry call, exactly as required; it also adds named regression tests for both the drop/keep cases, the omitted-field case, and the telemetry payload.

Review context

• Author: kai392
• Role context: outside_contributor
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: Python, Cuda, JavaScript, Kotlin, MDX, Perl, Ruby, TypeScript
• Official Gittensor activity: 95 PR(s), 7 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

[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 79.37%. Comparing base (8b44697) to head (619b68c).
⚠️ Report is 1 commits behind head on main.

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #9908       +/-   ##
===========================================
- Coverage   91.76%   79.37%   -12.39%     
===========================================
  Files         921      283      -638     
  Lines      113207    58868    -54339     
  Branches    27255     8718    -18537     
===========================================
- Hits       103886    46729    -57157     
- Misses       8034    11844     +3810     
+ Partials     1287      295      -992     

Flag	Coverage Δ
backend	95.69% <100.00%> (+<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-miner/lib/discover-cli.ts	94.86% <100.00%> (-5.14%)	⬇️
...kages/loopover-miner/lib/discovery-index-client.ts	100.00% <100.00%> (ø)

... and 771 files with indirect coverage changes

[loopover-orb]

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