Model routing sees what reaches the model. Cursor Agent Hooks see what the agent does on the developer machine. Together they cover the request and the work that follows it.
By Munis Badar · ·
An AI coding agent can ask a model for a plan, then read files, run a shell command, call an MCP server, or write a file locally. A gateway that sees only model traffic cannot refuse the command before it runs. Cursor hooks add that second control point while keeping policy in AI-FW.
The integration applies the same rules and tool governance model to Cursor events, with observe-first behavior so teams can understand usage before enabling blocking.
Enable Cursor Agent Hooks from AI Firewall -> Agent Config -> Integrations, begin with capture and observe-only governance, and review the Skills & Tools and Artifacts pages before selecting a blocking posture. This avoids treating an empty allow-list as an accidental deny policy.
The public hook package includes a hooks.json template plus Node and Python shims. The detailed guide explains scopes, credentials, repeat-window deduplication, failure posture, and privacy boundaries.