AgentSkills Specification
Official Agent Skills specification and documentation repo. Defines the SKILL.md format used by Claude Code, Codex, Copilot, and 20+ other AI coding tools.
- Create portable skill files that work across different coding assistants
- Publish a reusable skill to the community marketplace
- Validate your skill format before sharing it with your team
The official AgentSkills specification defining the SKILL.md format used by Claude Code, Codex, Copilot, and 20+ AI coding tools. 14.7K stars. The standard behind the ecosystem.
Skill authors and tool builders who want to create portable skills that work across multiple AI coding assistants.
https://github.com/agentskills/agentskills
By agentskills
How to Get It
This is a methodology or approach. Paste the instructions below into a Claude conversation to get started.
claude plugins install agentskills/agentskills
Tip: Paste this into a Claude Code conversation. Verify command matches your Claude Code version.
Trust Signals Reviewed
Community Pulse Active
Discussed on Hacker News, Reddit
- Agent Skills is now an open standard — Reddit · 558 pts
- Agent Skills — Hacker News · 544 pts
- SkillsBench: Benchmarking how well agent skills work across diverse tasks — Hacker News · 364 pts
31 mentions across 2 sources
Reviewer notes
Reviewed review. These are observations, not a security certification.
The spec itself — defines how skills work across the ecosystem. Foundational reference.
2026-05-10: Useful if you're building skills or agent workflows that need to run across multiple AI coding tools without rewriting instructions per platform—SKILL.md gives you a common format that Claude Code, Copilot, and others nominally support. The catch: this is a spec repo, not a runtime, so portability is only as good as each tool's actual implementation, and that varies. Check whether your specific target platforms have shipped SKILL.md execution support before betting a workflow on it; 15K stars signals ecosystem momentum, but production adoption (7 mentions) is still thin.
How to evaluate tools before deploying →
Data shown here comes from public APIs and automated scanning. Reviewer notes reflect one person's experience. This is not a security certification or legal recommendation. Always evaluate tools according to your own organization's policies.
Evaluation
The spec itself — defines how skills work across the ecosystem. Foundational reference.