How Skills Work
When Cursor starts, it automatically discovers skills from skill directories and makes them available to Agent. For a fuller introduction, see what are skills in Cursor. The agent is presented with available skills and decides when they are relevant based on context.
Skills can also be manually invoked by typing / in Agent chat and searching for the skill name, or by typing @ and selecting the skill to attach as context.
Skill directories
Skills are automatically loaded from these locations:
| Location | Scope |
|---|---|
.agents/skills/ | Project-level |
.cursor/skills/ | Project-level |
~/.cursor/skills/ | User-level (global) |
For compatibility, Cursor also loads skills from Claude and Codex directories: .claude/skills/, .codex/skills/, ~/.claude/skills/, and ~/.codex/skills/.
Each skill should be a folder containing a SKILL.md file:
.agents/
└── skills/
└── my-skill/
└── SKILL.md
Viewing skills
To view discovered skills:
- Open Cursor Settings (Cmd+Shift+J on Mac, Ctrl+Shift+J on Windows/Linux).
- Navigate to Rules.
- Skills appear in the Agent Decides section.
How do I use a skill?
- Type / followed by the skill name in chat to run it (e.g.
/write-tests). - Type @ and select the skill to attach it as context.
- Agent reads the skill file and follows the instructions.
To create your own skills, read creating skills in Cursor. To see why skills matter, read why we need skills in Cursor.
Summary
Skills are discovered from directories at startup. Invoke with /skill-name or @skill-name, or let the agent apply them automatically based on context.