> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.merge.dev/merge-agent-handler/build/skills/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.merge.dev/_mcp/server.

# Skills

> Build once. Connect to every API, tool, and LLM. Ship products and agents to production without owning the integration stack.

## Docs

- [Skills](https://docs.merge.dev/merge-agent-handler/build/skills/overview.md): What a skill is in Agent Handler, how Skills serves them to the AI clients your employees use, and how skills relate to tools and connectors.
- [Writing a skill](https://docs.merge.dev/merge-agent-handler/build/skills/writing-a-skill.md): How to author a skill for Agent Handler: SKILL.md frontmatter, bundled reference files, size limits, validation rules, and authoring patterns that work.
- [Publishing skills](https://docs.merge.dev/merge-agent-handler/build/skills/publishing-skills.md): How to publish a skill in Agent Handler: the upload and preview flow, publishing to the org, a group, or yourself, drafts, versions, unpublishing, and enabling Merge-provided catalog skills.
- [Reviewing submissions](https://docs.merge.dev/merge-agent-handler/build/skills/reviewing-submissions.md): How employee-authored skills work in Agent Handler: submitting for review, approving or rejecting a submission, and the resubmit loop.
- [How agents load skills](https://docs.merge.dev/merge-agent-handler/build/skills/how-agents-load-skills.md): How agents discover and retrieve skills in Agent Handler: the list_skills and retrieve_skill MCP tools, progressive disclosure of bundled files, and retrieval auditing.