rule-validation
Validates Skill, Agent, and Command syntax using validate_skills.py, logs errors, and manages the automated QC workflow for agent development.
Introduction
The rule-validation skill is a critical component of the agent generation framework, designed to ensure the integrity and quality of newly generated AI agents. It acts as an automated gatekeeper that enforces syntactical correctness for Skills, Agents, and Command definitions within the repository. By leveraging the internal validate_skills.py script, this tool systematically scans designated directories to identify configuration errors, structural inconsistencies, and missing documentation. It serves developers and technical users who require a rigorous, repeatable process for maintaining high-quality agent outputs, ensuring that all components meet the framework's strict architectural standards before deployment.
-
Executes comprehensive syntax validation for .claude/skills/ repositories and generated agent structures.
-
Integrates with the validation_checklist.md to ensure all prerequisites are met before formal assessment.
-
Automates the creation of skills_check_log.md to maintain a clear audit trail of validation attempts, error logs, and subsequent fixes.
-
Implements a mandatory quality control (QC) loop, delegating advanced evaluations to specialized subagents like qa-skill-qc.
-
Ensures that workflows strictly adhere to validation_criteria.md, promoting consistency across diverse domain-specific agent projects.
-
Prerequisites: Always review the validation checklist and ensure all SKILL.md files are present before triggering the script.
-
Workflow: The validation process is iterative. Users must repeat the execution of validate_skills.py until the All skills passed validation message is received.
-
Logging: All error findings, specific fix actions taken, and rationale behind corrections must be documented in the check log to maintain provenance.
-
Collaboration: The skill relies on the subagent_policy, requiring that quality loops and feedback cycles be handled by recommended subagents.
-
Constraints: The process mandates a maximum of three refinement cycles based on QC feedback to balance speed with development rigour.
Repository Stats
- Stars
- 10
- Forks
- 7
- Open Issues
- 0
- Language
- Python
- Default Branch
- main
- Sync Status
- Idle
- Last Synced
- May 3, 2026, 04:52 PM