fix: close the remaining 5 pre-existing test failures on main

[JSONbored]

Summary

Follows #9276, which fixed 8 files / 12 cases and left 4 files / 5 cases still red on main — and therefore red on every open PR, including unrelated ones like #9283. Each traced to its introducing commit. Two are real source gaps; two are stale sibling tests that a deliberate behavior change never updated.

Source fixes

src/selfhost/metrics.ts — registers two counters emitted since #9243 (loopover_orb_relay_multiple_live_enrollments_total) and #9245 (loopover_private_manifest_warnings_total) but never added to DEFAULT_METRIC_META. Without an entry, renderMetrics() emits them as bare undocumented samples with no # HELP/# TYPE — exactly the gap that drift guard exists to catch. Both commits name the new counter in their own body text, so this is a pure registration miss, not a behavior decision.

Test fixes (source verified correct in both cases)

selfhost-pg-retention — the fake Postgres still matched the pre-#9083 ctid semi-join shape, so every DELETE fell through to rowCount: 0 and nothing was ever counted as deleted. #9083 deliberately moved retention to an index-backed PK range delete with ORDER BY — the fix for prune-retention blowing its 30-minute timeout and stalling retention permanently. It updated the SQLite twin (retention.test.ts) but not the Postgres one. I verified the source is correct before touching the test: all 14 mapped PK columns are real single-column primary keys, the WHERE predicate is unchanged so the eligible set is identical, and ORDER BY only makes batches deterministically oldest-first. Retention deletes neither too much nor too little. Regex now captures whichever key column is in play, so both the mapped-PK and ctid-fallback paths stay exercised.

salvageability — #9085 made an absent blocker confidence degrade to CONFIDENCE_WHEN_UNSTATED (0.5) instead of 1.0, so it is sub-floor against the 0.93 gate default and the low-confidence hold owns that case. #9085 renamed both sibling assertions in rules.test.ts but missed this third consumption site. Both calls kept deliberately — they are the only coverage in that file of the aiReviewCloseConfidence ?? DEFAULT_AI_REVIEW_CLOSE_CONFIDENCE default arm and the confidence ?? CONFIDENCE_WHEN_UNSTATED nullish arm respectively.

worker-entry-boundary — a false positive, not a real dependency leak. The check grepped whole-file text, so #9230's user-facing string "route(s) crossed the visual-diff threshold" failed a green tree over a sentence — in a file worker-reachable since #4120 that imports none of these deps. A Node-only dep can only reach the Worker bundle by being imported, so the check is now scoped to module specifiers, reusing the file's own existing parseImportSpecifiers. Bending correct user-facing copy to dodge a test regex would have been the wrong repair.

Test plan

• All 4 originally-failing files green: npx vitest run test/unit/salvageability.test.ts test/unit/selfhost-metrics.test.ts test/unit/selfhost-pg-retention.test.ts test/unit/worker-entry-boundary.test.ts — 69 passed
• Neighbouring/sibling suites still green: retention.test.ts, selfhost-pg-dialect.test.ts, alerts-metric-name-references.test.ts (builds its allowed set from the same array), rules.test.ts, selfhost-pg.test.ts — 191 passed, 11 skipped
• npx tsc --noEmit -p tsconfig.json — clean
• Verified the narrowed boundary check is still discriminating, not vacuous: injected a real import sharp from "sharp" into a worker-reachable file → check failed, naming the exact file and specifier; removed it → green. Also added a permanent test pinning both directions (flags real specifiers, ignores prose).

[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 88.54%. Comparing base (f7dda79) to head (50c8d6c).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #9285       +/-   ##
===========================================
+ Coverage   75.38%   88.54%   +13.16%     
===========================================
  Files         275      836      +561     
  Lines       58023   109515    +51492     
  Branches     6181    26084    +19903     
===========================================
+ Hits        43739    96971    +53232     
+ Misses      14014    11562     -2452     
- Partials      270      982      +712     

Flag	Coverage Δ
backend	93.54% <ø> (?)

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

Files with missing lines	Coverage Δ
src/selfhost/metrics.ts	100.00% <ø> (ø)

... and 698 files with indirect coverage changes

[loopover-orb]

Warning

⏸️ LoopOver review result - manual review recommended

_{Review updated: 2026-07-27 12:10:52 UTC}

4 files · 1 AI reviewer · 2 blockers · CI green · clean

⏸️ Suggested Action - Manual Review

Review summary
This PR registers two previously-unregistered metric names in DEFAULT_METRIC_META (metrics.ts) and updates three stale test files to match deliberate behavior changes from prior PRs (#9083 retention delete shape, #9085 confidence-degrade default, #9230 worker-boundary check narrowed to import specifiers rather than raw file content). Each test fix is traced to a specific prior commit with a clear rationale, and the salvageability test now correctly asserts `toBeUndefined()` where it previously (incorrectly) asserted `toBeDefined()` post-#9085. The worker-entry-boundary regex change (scanning import specifiers instead of raw content) is a legitimate fix for a real false-positive class (prose containing forbidden words), and the added self-test for the regex's discriminating power is a reasonable coverage addition, not fabricated.

Nits — 4 non-blocking

• metrics.ts: the two new HELP strings are quite long (multi-clause with inline issue refs) compared to most existing entries — consider trimming for readability, though this is consistent with a few other recent entries in the file.
• worker-entry-boundary.test.ts: the new self-test duplicates literal specifier strings ('sharp', 'pixelmatch', etc.) already encoded in FORBIDDEN_IDENTIFIERS — fine for a meta-test but worth a one-line comment noting it's intentionally redundant with the regex definition.
• Consider adding a short changelog-style comment near DEFAULT_METRIC_META noting that new counters must be registered here at introduction time, to prevent the next unregistered-counter drift the PR description flags.
• No further changes needed for merge from a code-correctness standpoint; verify only that the linked issue-closing convention is satisfied per repo policy.

Concerns raised — review before merging

• No linked issue detected: No closing reference or linked issue number was found in the PR metadata/body. — If this PR is intended to solve an issue, link it explicitly in the PR body.
• Maintainer requires a linked issue: This repo's maintainer focus manifest requires every PR to reference a tracked issue. — Link the relevant issue (for example Closes #123) before opening the PR.

📋 Copy for AI agents — paste into your coding agent

Fix the following blocker(s) from this PR review:

1. No linked issue detected: No closing reference or linked issue number was found in the PR metadata/body. — If this PR is intended to solve an issue, link it explicitly in the PR body.

2. Maintainer requires a linked issue: This repo's maintainer focus manifest requires every PR to reference a tracked issue. — Link the relevant issue (for example `Closes #123`) before opening the PR.

Decision drivers

• ❌ Code review — 2 blockers (1 reviewer)
• ❌ Gate result — Blocking (Repo-configured hard blocker found.)

Context & advisory signals — never blocks the verdict

Signal	Result	Evidence
Linked issue	⚠️ Missing	No linked issue or no-issue rationale found.
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 (no linked issue context).
Validation posture	✅ 25/25	PR body includes validation/test evidence.
Contributor workload	✅ 10/10	Author activity: 13 registered-repo PR(s), 13 merged, 307 issue(s).
Contributor context	✅ Confirmed Gittensor contributor	JSONbored; Gittensor profile; 13 PR(s), 307 issue(s).
Improvement	✅ Minor	risk: clean · value: minor · LLM: moderate

Review context

• Author: JSONbored
• Role context: owner (maintainer lane)
• Public audience mode: oss maintainer
• Lane context: Repository is configured for direct PR review.
• Public profile languages: not available
• Official Gittensor activity: 13 PR(s), 307 issue(s).
• PR-specific overlap: none found.

Contributor next steps

• Start here: Treat this as maintainer-lane context rather than normal contributor-lane activity.
• Then work through the remaining 2 steps in the Signals table above.

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