Sentry
Pulls real production errors and stack traces into the agent's context so it can debug against what actually broke.
What it does
Gives an agent access to your Sentry issues: error details, stack traces, affected releases, and event frequency. Maintained by Sentry.
The workflow it enables is worth the setup. Instead of pasting a stack trace into a chat, you point the agent at the issue and it reads the trace, the breadcrumbs, and the release context itself.
Setup
The hosted remote server handles auth through OAuth in the browser:
claude mcp add --transport http sentry https://mcp.sentry.dev/mcp
What to watch out for
Error payloads can contain user data. If your events carry request bodies, emails, or tokens, connecting this pipes that into your model provider. Check what your Sentry scrubbing rules strip before wiring it up.
Useful mainly when you already run Sentry. It is not a reason to adopt Sentry.
