Skip to content

Agent Skills

Agent Skills Directory

Agent skills are reusable workflows that help AI agents perform specialized tasks. Skills Directory focuses on skills that are discoverable, source-attributed, and security-scanned before installation.

Indexed skills

592,677

Search focus

agent skill

Security context

Scanned

Top matching skills

Skills vs tools

Agent skills usually define how an agent should reason, inspect, write, test, or operate. Tools and MCP servers expose actions; skills turn those actions into repeatable workflows.

Cross-agent workflows

Many SKILL.md-style workflows can be adapted across Claude Code, Cursor, Codex, OpenCode, and other coding agents. The most portable skills are text-first and avoid hardcoded local assumptions.

Quality signals to check

Look for clear triggers, narrow scope, maintained source repos, good examples, minimal hidden behavior, and a security grade that matches your risk tolerance.

Why security-scanned skills matter

Skills can influence agent behavior, suggest commands, include helper files, or touch code and data. Skills Directory adds security context so you can inspect risk before installing a workflow from a public repo.

See the security model

Frequently asked questions

What is an agent skill?
An agent skill is a reusable package of instructions, examples, and sometimes helper files that guides an AI agent through a specific task or workflow.
Do agent skills work outside Claude?
Some do. SKILL.md-style skills are most native to Claude/Claude Code, but many workflows can be adapted to Cursor, Codex, OpenCode, and other agent systems.
How many agent skills should I install?
Install only the skills you actively use. Too many overlapping skills can create conflicting instructions and make agent behavior less predictable.