[eslint-refiner] ESLint Refiner — Daily Report (2026-10-05): open-issue queue hits zero

[github-actions[bot]]

ESLint Refiner — Daily Report (2026-10-05)

Headline: the open-issue queue hit zero for the first time. All 3 issues that were open as of yesterday's run closed completed in the last ~24 hours, and each fix was independently re-verified against current rule source (not just trusted from bot state). No new issues filed today — a disciplined, clean-slate run.

Key metrics

Metric	Value
Rules tracked	65 (+1 since last run — no-unguarded-async-timer-callback)
Open eslint-factory issues	0
Issues closed completed (confirmed fixed) since last run	3
New issues filed today	0
Rules reviewed today	2 (1 first-ever review, 1 second-pass)

Confirmed fixes (source-verified, not just bot state)

• no-github-request-interpolated-route: fallback-expression Octokit client aliases (x || github, ternary) bypass client detection #64186 no-github-request-interpolated-route — fallback-expression Octokit client aliases (x || github, ternaries) now correctly detected. isOctokitSourceExpression gained LogicalExpression/ConditionalExpression recursion exactly covering the reported gap.
• require-escaped-regexp-interpolation: no recognition for regex-fragment constants (recurrence of #63566) #65515 / require-escaped-regexp-interpolation: no recognition for regex-fragment constants (5 sites already need manual disable, 1 unsupp #63566 require-escaped-regexp-interpolation — regex-fragment constants (e.g. AWF_API_PROXY_HOST_RE_SOURCE) are no longer misflagged as unescaped. A new naming-convention pattern (_PATTERN/_RE/_RE_SOURCE/_CHAR/_DELIMITERS suffix) plus one-hop const-initializer resolution now recognizes them. Verified against the exact live site in handle_agent_failure.cjs:72. Theme retired — do not refile.
• no-unsafe-catch-error-property: rename alias of catch param bypasses unsafe-property detection #65255 no-unsafe-catch-error-property — one-hop rename-alias resolution (const err = error; then err.message) confirmed working, consistent with the 10-04 finding.

Today's rule reviews (no new findings)

no-unguarded-async-timer-callback (new, first-ever review)
Flags async callbacks passed to setTimeout/setInterval/setImmediate unless the entire body is a single guarding try/catch. Only one live call site exists in the whole actions/setup/js corpus: process_runner.cjs:458-485 (a setInterval(async () => {...}) runtime-guard poller).

Found a real but untested structural asymmetry: the rule's isFullyGuarded check requires the callback body to be exactly one statement (the try). The test suite explicitly confirms trailing code after the try should be flagged (correct — it can throw and reject). But the same check also rejects a plain leading guard-clause return (if (done) return;) before the try, even though a boolean early-return cannot itself cause a rejection.

The one live site does have two such leading guards — but it also has genuinely risky unguarded code after the try (terminateTree(...), clearInterval(...), log(...)), so the rule's current flag there is still a legitimate true positive overall. Not a clean, isolated false-positive demonstration — not filed. Will revisit if a second, cleaner call site appears.

no-core-error-then-process-exit / no-core-error-then-process-exitcode (2nd pass)
Checked a fresh angle: named-constant or variable exit codes (process.exit(EXIT_FAILURE)) after core.error(...), since the rule only recognizes literal non-zero integers. Grepped ~30 process.exit(/process.exitCode = sites corpus-wide — none pair with core.error (all use console.error/process.stderr.write/a logger instead). Zero grounding, 2nd consecutive clean check (first was 09-14) — demoted further in rotation.

Next actions

• Keep watching for new issues landing between runs now that the queue is empty — the dedupe check still runs every day.
• no-unguarded-async-timer-callback: only refile the leading-guard asymmetry if a second, cleaner live call site appears (guard + try + nothing risky after).
• Chronic, unactionable because this workflow has no create_pull_request safe output: the shared try-catch-rule-utils.ts VariableDeclaration-suggestion gap (~14 rules affected, twice filed and expired as eslint-factory: shared try/catch suggestion builder silently skips VariableDeclaration call sites #57868/eslint-factory: shared try/catch suggestion builder still skips VariableDeclaration call sites -- recurrence of expired issue 57 #59891) — report-only, not refiling again.

Strategy and full history persisted to repo-memory (memory/eslint-refiner branch) for continuity.

Generated by 🤖 ESLint Refiner · claude · agent · 255 AIC · ⌖ 6.81 AIC · ⊞ 5.1K · ◷

• expires on Oct 5, 2026, 9:45 PM UTC-08:00

[github-actions[bot]]

This discussion was automatically closed because it expired on 2026-10-06T05:45:06.192Z.

Closed by Workflow

[github-actions[bot]]

This discussion was automatically closed because it expired on 2026-10-06T05:45:06.192Z.

Closed by Workflow