TOOLID=human-review-queue-auditor. A zero-dependency Node 22+ reporter for exported queue snapshots. It never approves, rejects, assigns, pauses, resumes, or modifies a review. It reads local JSON only and emits prioritized items with fixed source-role provenance, not raw IDs or owner names.
This walkthrough uses the tool's public README and checked-in example files. Run the command from a repository checkout with Node.js 22+; inspect the source before using it on your own files.
Run the checked-in example
node bin/human-review-queue-auditor.mjs --root examples/passing --policy policy.json --queue queue.json
node bin/human-review-queue-auditor.mjs --root examples/failing --policy policy.json --queue queue.jsonRead the result
The first example exits 0 and the second exits 1 with stale-unowned at /items/0 in @queue. @queue and @policy are logical source roles for the exact files named at invocation, not host paths.
Where this check stops
Queue 1,048,576 bytes; policy 65,536 bytes; JSON depth 16; 1,000 items; 5,000 transitions; evaluation time 5,000 ms with injected clock. Bounds are inclusive; N+1 is incomplete for the queue or invalid configuration for the policy. Input is strict UTF-8, realpath-confined to --root; the tool writes nothing and makes no network calls. No live queue API, notification, assignment, auto-approval, retry, or workflow mutation is included. The library exports TOOLID, LIMITS, RULES, validPolicy and auditQueue(queue,policy,{now,deadline}).
Before adapting the command to your own workflow, review the accepted inputs, exit codes and safety boundaries in the README.
Compiled with AI assistance from checked-in public documentation and example scripts. Run the example and review the repository's current documentation before relying on its result.