Integrations for Claude Codes
Claude Code integrations connect the agent to outside tools and to your own workflow events. The Model Context Protocol, MCP, lets it call servers that expose data and actions such as issue trackers or databases, while hooks run your commands at lifecycle points like before or after a tool call to enforce rules.
Stage: Extending · Part of the Claude Code roadmap · Last reviewed 19 November 2026
What to learn in Integrations
- Add an MCP server and list the tools it provides
- Scope an MCP server to a project or a user
- Write a hook that runs a formatter after file edits
- Block a risky tool call with a pre-tool-use hook
- Judge the trust and permission impact of a third-party server
Integrations interview questions
What is MCP in Claude Code?
The Model Context Protocol is an open standard for connecting AI tools to external systems. An MCP server exposes tools and resources, such as querying a database or reading tickets, and Claude Code can call them as if they were built in.
How is a hook different from a prompt instruction?
A prompt instruction is advice the model may or may not follow. A hook is a command run by the harness at a defined event, so it executes deterministically every time, which makes it suitable for formatting, validation and policy checks.
Where Integrations fits in the Claude Code roadmap
Integrations belongs to the Extending stage (beginner level), which takes about 6 hours. Tools and agents.
Other units in this stage
Back to the full Claude Code roadmap