Agent-Skills.md

Agent Skills: Skill: critic-run

Run TOOLS/governance/critic.py to verify governance compliance before making changes

UncategorizedID: Reneromero08/agent-governance-system/critic-run

Repository

Reneromero08/agent-governance-system
Reneromero08License: NOASSERTION

Install this agent skill to your local

pnpm dlx add-skill https://github.com/Reneromero08/agent-governance-system/tree/HEAD/CAPABILITY/SKILLS/governance/critic-run

Skill Files

Browse the full folder contents for critic-run.

Download Skill

Loading file tree…

CAPABILITY/SKILLS/governance/critic-run/SKILL.md

Skill Metadata

Name
critic-run
Description
Run TOOLS/governance/critic.py to verify governance compliance before making changes

required_canon_version: >=3.0.0

Skill: critic-run

Version: 0.1.0

Status: Active

Purpose

Runs the governance critic tool to verify that proposed changes comply with AGS governance rules. This is a critical pre-commit check that ensures canon integrity, contract compliance, and architectural decision alignment.

Trigger

Use when:

  • Before committing changes to the repository
  • When validating governance compliance during CI
  • As part of the commit ceremony workflow

Inputs

JSON object with optional fields:

  • verbose (boolean, optional): Enable verbose output (default: false)

Outputs

JSON object:

  • passed (boolean): Whether all governance checks passed
  • output (string): Combined stdout and stderr from critic.py
  • exit_code (integer): Process exit code (0 = pass)

Constraints

  • Read-only operation (does not modify files)
  • Must be run from repository root
  • Requires Python 3.8+

Implementation

Wrapper around CAPABILITY/TOOLS/governance/critic.py that:

  1. Executes the critic script via subprocess
  2. Captures output and exit code
  3. Returns structured JSON result

Fixtures

  • fixtures/basic/ - Basic invocation test

required_canon_version: >=3.0.0