The unbiased chart for
agent skills.

Skills are reusable capabilities for AI agents that provide procedural knowledge for specific tasks like code generation, domain workflows, and tool integrations.

A .skill file is essentially a zip file with a .skill extension. You can inspect it by extracting the archive and browsing the folders inside.

Skills help agents complete specific tasks more effectively by giving them structured instructions and reusable patterns. They're like tutorials for your AI agents, so you don't have to explain things over and over.

Downloaded skill files should never be added blindly. Review the instructions, scripts, and templates before importing anything into your agent environment.

A skill typically includes a SKILL.md instruction file and can also include optional scripts and templates.

Skills can be model-invoked, meaning the agent can automatically decide when to use them based on the task and context.

No, skills are increasingly getting more broadly used as reusable capabilities for AI agents, not just coding add-ons.

Skills are compatible with most AI agents like Openclaw, Manus, ChatGPT, Codex, Cursor, Antigravity, Claude, Perplexity Computer, and Minimax.

Every SKILL.md starts with a YAML frontmatter block between --- markers at the top. It contains metadata like name and description, which agents and directories can use to understand when the skill applies.

The name field must be lowercase, use hyphens instead of spaces, can include digits, and maxes out at 64 characters. Think of it like a URL slug: my-cool-skill works, My Cool Skill! does not.

The description should explain both what the skill does and when it should be triggered. AI agents use the description to decide whether to invoke the skill, so vague descriptions can cause missed or incorrect usage.

Yes, some advanced skill bundles include multiple SKILL.md files for related sub-skills, each with its own instructions and metadata.