Coordinate and enforce cross-harness discipline for agent skills and infrastructure.
npx skills add https://github.com/misty-step/volume --skill harness-misty-step
Please help me install this Agent Skill. Skill: harness Source: https://github.com/misty-step/volume/tree/main/.agents/skills/harness Command: npx skills add https://github.com/misty-step/volume --skill harness-misty-step
Build, maintain, evaluate, and enforce the agent harness — skills, agents, hooks, CLAUDE.md, AGENTS.md, and enforcement infrastructure. Use when: "create a skill", "update skill", "improve the harness", "sync skills", "eval skill", "lint skill", "tune the harness", "add skill", "remove skill", "convert agent to skill", "audit skills", "skill health", "unused skills", "lint this skill", "eval this skill", "evaluate this skill", "validate this skill", "check this skill".
Run the setup to initialize the harness scaffolding and begin experimenting with skills and agents.
High-intent search queries and answers about installing and using this skill.
To audit and lint agent skills across multiple projects, you enforce cross-harness discipline for hooks, CLAUDE.md, and AGENTS.md. This coordinates governance workflows to ensure consistent behavior and safe distribution across Claude Code, Codex, and Pi environments.
An agent harness encompasses your skills, agents, hooks, and configuration files like CLAUDE.md. You need to enforce harness governance when building, evaluating, or syncing these components to maintain quality gates and ensure safe lifecycle distribution across projects.
To convert an agent into a reusable skill, you route its modes, references, and hooks through centralized harness governance. This encodes design principles and lifecycle enforcement to ensure the resulting skill scales safely across different project environments.
Cross-harness skill governance works with Claude Code, Codex, and Pi. It provides centralized routing of modes, references, and hooks to ensure scalable workflows and compatible behavior across these distinct platforms.
You check for unused or unhealthy skills by running a harness audit to evaluate skill health and lifecycle state. This validates your skills against design principles and quality gates to identify components safe for removal or update.
Manually syncing skills without harness governance risks inconsistent behavior and breaking quality gates across projects. Without centralized enforcement of lifecycle hooks and design principles, safe distribution across environments like Claude Code and Codex cannot be guaranteed.
cdrguruInitialize Gemini CLI configuration and auditor persona for the PACK workflow.
Adonis0123Migrate Claude skill-sync workflows between repositories with scripts and hooks.
Poorgramer-ZackDesign, validate, and test Copilot CLI hooks for policy enforcement.
asfbay-bitAudit code quality with Auditor, Fixer, and Verifier agents.
