Sentry MCP server
Triage Sentry issues, search events, and pull Seer root-cause analysis.
About
Connect Tracecat to Sentry to triage errors and incidents from agents working alongside SREs and on-call engineers. You can search issues, pull the failing event with full stack traces, and request a Seer root-cause analysis without an engineer touching the dashboard. From there, agents can resolve or ignore benign issues, assign owners, and push the enriched context into a Linear ticket or a Tracecat case so the on-call engineer stays focused on the long tail.
Setup
- 1
Sign in with OAuth
You'll authorize Tracecat to access Sentry on your behalf. No API keys to manage.
- 2
Select the
Sentrytile in the Tracecat MCP catalogOpen the
MCP catalogin your workspace, select theSentrytile, and complete the OAuth flow. - 3
Enable
Sentryin your agentIn your
ai.agentaction orAgents→toolstab, selectSentryfrom theMCP integrationsdropdown.
Tools
find_organizations | List Sentry organizations the connected user can access. |
find_projects | List projects in a given organization with platform and team info. |
search_issues | Natural language search across Sentry issues with AI-translated queries. |
get_issue_details | Fetch a single issue with stack traces, breadcrumbs, and tags. |
search_events | Search events using natural language translated to Sentry's query syntax. |
update_issue | Resolve, ignore, assign, or comment on a Sentry issue. |
analyze_issue_with_seer | Trigger or fetch a Seer AI root-cause analysis for an issue. |
create_dsn | Create a new client key (DSN) for a project. |
find_releases | List releases for a project with deploy info and commits. |