Sentry
Let agents inspect production errors and Sentry issue context.
Sentry MCP lets agents query your Sentry account directly. Agents can look up real errors with stack traces, request data, releases, and user context.
What agents can do
Once you connect Sentry, agents can:
- Find recent issues in your Sentry organizations and projects
- Read stack traces and event details
- Compare releases and suspect commits
- Triage issues: assign teammates, resolve, and update status
- Use production error context while fixing bugs
Setup
Go to Workspace Settings → Connectors (or a project's Connectors page) and click Connect on the Sentry card (or Connect another to add another account). You can connect your own Sentry account over OAuth, or add an access token.
Connect your own Sentry account
- Choose who the connection belongs to and where agents can use it. See connection options for details.
- Click Continue, then Continue to Sentry. You can select the individual permissions to grant on Sentry's approval screen.
Agents you launch then use Sentry's hosted MCP server with your own access, so they see exactly the organizations and projects you can.
If Sentry stops accepting the saved connection, the Connectors page shows a Reconnect needed badge on the connection, and agents you launch run without Sentry until you reconnect.
Access token
Generate a token at sentry.io/settings/account/api/auth-tokens.
Recommended scopes for full Sentry MCP compatibility:
org:read, team:read, project:read, project:releases, event:read, team:write, project:write, event:write- Fill in the connection details as above, then click Continue.
- Open the Access token tab, paste your Sentry Access Token, and click Connect Sentry.
When to use it
Ask agents to use Sentry when a ticket references a production error, stack trace, crash, affected user, or release regression.