consensus-loop:audit

Review pending trigger items in the watch file to audit the consensus-loop.

15|4|Updated Mar 15, 2026
One-click install
npx skills add https://github.com/berrzebb/consensus-loop --skill consensus-loop-audit
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: consensus-loop:audit
Source: https://github.com/berrzebb/consensus-loop/tree/main/skills/audit
Command: npx skills add https://github.com/berrzebb/consensus-loop --skill consensus-loop-audit

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill enables teams to perform a manual audit of the consensus-loop process by inspecting pending trigger_tag items in the watch file, ensuring audits occur before merging.

Core Features & Use Cases

  • Manual audit trigger: Initiate an audit without editing the watch file.
  • Audit testing: Re-test a failed audit or validate the audit prompt.
  • Clear outcomes: Produce a verdict document detailing results and next steps.

Quick Start

Run the audit using the audit.mjs script with your chosen arguments.

Frequently Asked Questions about consensus-loop:audit

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I manually trigger a consensus-loop audit without editing the watch file?

You can trigger a manual consensus-loop audit by running the audit script directly, which inspects pending trigger items in the watch file without requiring you to modify it.

What is a manual audit of consensus-loop trigger items used for?

A manual audit of consensus-loop trigger items reviews pending entries in the watch file, ensuring verification checks occur before merging and producing a verdict document with next steps.

Can I re-run a failed consensus-loop audit or test the audit prompt?

Yes, you can re-run a failed consensus-loop audit or validate the audit prompt by executing the audit script, which allows you to test and verify the process independently.

Does the consensus-loop audit support a dry run to test behavior without applying changes?

The consensus-loop audit supports a --dry-run CLI option to test audit behavior safely, alongside other flags like --no-resume, --auto-fix, and --model to control execution.

Why does my manual audit of consensus-loop items not produce a clear verdict?

If your manual audit of consensus-loop items does not produce a clear verdict, ensure the watch file contains pending trigger items and verify your CLI options are correctly configured for the audit script.

What is the best way to control the model used during a consensus-loop audit?

The best way to control the model during a consensus-loop audit is to specify it using the --model CLI option when running the audit script, allowing you to direct the verification process.