Skill Governance Balanced

What This Skill Does

The Skill Governance Balanced agent is a comprehensive management framework designed to oversee the lifecycle of OpenClaw AI skills. Developed by dalomeve, this tool provides a structured environment for maintaining skill health, preventing side-effect proliferation, and optimizing the selection of tools during agent execution. It functions as a gatekeeper that enforces a robust status model: candidate, ready, core, quarantine, and retired. By tracking usage metrics and validation outcomes, it ensures that only high-quality, verified skills are utilized in your agent's primary decision-making path.

Installation

To integrate this governance layer into your workspace, execute the following command in your OpenClaw terminal:

clawhub install openclaw/skills/skills/dalomeve/skill-governance-balanced

Once installed, you must initialize the workspace state. Navigate to the root directory and run the reconcile-ready.ps1 script. Before any new skill can be utilized for production tasks, it must pass the 4-step acceptance audit by running scripts/audit-skill.ps1 -Root <workspace> -SkillName <name>. Only when the skill status is updated to ready in skill-registry.json is it cleared for use.

Use Cases

This skill is ideal for power users and developers managing large libraries of OpenClaw tools. Key use cases include:

• Automated Lifecycle Management: Automatically promoting frequently used skills to the 'core' tier while demoting stale ones to keep the selection pool optimized.
• Stability Control: Quarantining skills that exhibit repeated failure patterns, preventing them from destabilizing the agent execution flow.
• Optimized Routing: Leveraging dynamic core-cap ranges (8 to 14) to balance the agent's decision speed against its breadth of available toolsets.
• System Maintenance: Utilizing the weekly cleanup script to prune orphaned or retired skill configurations without deleting actual source files, ensuring a tidy registry.

Example Prompts

1. "Run the governance reconcile script and let me know which of my currently installed skills are ready for production use."
2. "Execute the audit process for the browser-automation skill to see if it qualifies for the ready status."
3. "Update the core pool configuration; what skills were demoted based on their usage data from the last three days?"

Tips & Limitations

• Governance is Local: This skill does not communicate with external servers. It relies exclusively on your local skill-registry.json file for its state machine.
• Manual Audit: The governance model is proactive but requires initial setup. Always ensure you run the audit script before expecting a new tool to appear in the route-skill logic.
• Avoid Manual Editing: While the registry is a JSON file, avoid manually editing it. Always use the provided PowerShell scripts to maintain data integrity and avoid registry corruption.
• Dynamic Range: Remember that the dynamic core cap is strictly enforced; if you exceed the limit, the system will automatically demote the least active core skills.