Skills and custom slash commands

The two used to be separate features. They have merged, and the exam-relevant details are the frontmatter and the precedence.

6 min read · Lesson 8 of 11 in this domain

A skill is a folder containing a SKILL.md file whose instructions Claude can load. The key economic property is that the body loads only when the skill is used — so a long procedure costs almost nothing until it is needed, unlike CLAUDE.md content which is in context from launch. That is the reason to move a grown-out CLAUDE.md section into a skill. Note the consolidation: custom commands have merged into skills. A file at .claude/commands/deploy.md and a skill at .claude/skills/deploy/SKILL.md both create /deploy and behave the same way; existing command files keep working, but skills add a directory for supporting files, frontmatter controlling who may invoke them, and automatic loading when relevant.

• Project scope: .claude/skills/<name>/SKILL.md. User scope: ~/.claude/skills/<name>/SKILL.md.
• disable-model-invocation: true stops Claude loading it automatically — use it for actions you want to trigger manually with /name.
• user-invocable: false hides it from the / menu, for background knowledge users should not run directly.
• context: fork runs the skill in its own subagent context, isolating it from the main conversation.
• allowed-tools scopes which tools the skill may use; description is what Claude reads when deciding relevance.
• Name conflicts resolve enterprise → personal → project, any level overrides a bundled skill, and where a skill and a command share a name the skill wins.
• Plugin skills are namespaced plugin-name:skill-name, so they cannot collide.
• Nested .claude/skills/ below the working directory become available when Claude reads a file in that subdirectory.

Source: Claude Code Docs — Skills · Independent study aid, not affiliated with or endorsed by Anthropic.