Sentry and Claude Code: connect the MCP server and triage errors daily

What Claude Code does for Sentry triage

• Reads unresolved Sentry issues that are new or regressed since its last run.
• Checks your board first, so a noisy error gets a comment instead of ten new issues.
• Files each new error with its impact, first and last seen, the release it started in, the top stack frames and a likely cause, marked as a guess.
• Fixes the small, obvious ones in a pull request with a test, and waits for review.

Claude Code features for Sentry triage

How to set up Claude Code for Sentry triage

1. Connect the tools

   Add Sentry's hosted server, then run /mcp and sign in with your Sentry account in the browser. It uses OAuth, so there's no token to copy.

   # Add Sentry's hosted MCP server
   claude mcp add --transport http sentry https://mcp.sentry.dev/mcp
   
   # Then, in a session, sign in with OAuth
   /mcp

2. Write the rules in CLAUDE.md

   Claude Code reads CLAUDE.md at the start of every session, claude -p runs included. Put the job's rules there, or in a skill under .claude/skills/ if they're long. See the CLAUDE.md guide.

   ## Sentry triage rules
   1. List unresolved production issues first seen or regressed
      since the last run.
   2. Search the board for the Sentry issue ID before filing.
   3. File new ones with users and events affected, first and last
      seen, the release, the top stack frames and a likely cause,
      marked as a guess.
   4. Urgent if it blocks sign-up, payment or data; High above 1%
      of sessions; otherwise Medium.
   5. Small, obvious fix: open a pull request with a test. Don't merge.
   6. Never resolve, ignore or delete anything in Sentry.

3. Run it without the interactive interface

   claude -p runs one task and exits, so cron, CI or another tool can start it. Choose what it may do without asking with --allowedTools or --permission-mode; --output-format json returns the result with its cost.

   # A triage pass from cron, weekdays at 9:00 and 14:00
   claude -p "Triage new Sentry errors with the rules in CLAUDE.md" \
     --allowedTools "mcp__sentry" --output-format json

4. Put it on a schedule

   Make it a cloud scheduled task with /schedule (Pro, Max, Ultra, Team and Enterprise), a desktop scheduled task, or /loop while a session stays open. See how to make Claude Code proactive.

Source: Claude Code documentation

Limits and safety

• It needs to read Sentry issues, not change them. Leave resolving and ignoring to people.
• Its cause is a guess from the stack trace. A person confirms it before a fix ships.
• Errors can carry user data. Keep the issues it files on a private board or repository.
• Since Claude Code v2.1.302, claude -p starts in Manual mode: an action that needs approval is denied unless a rule allows it. Allow only the tools the job needs.

Claude Code or Codex for Sentry triage?

The rules carry over. Write them once in AGENTS.md and put @AGENTS.md in CLAUDE.md, and both agents read the same file.

Step by step for the other agent: Sentry triage with Codex.

Make Sentry triage someone's job

Commands run when someone remembers to run them. Crewmark turns your own Claude Code into an AI employee with a role, a goal in numbers, working hours and a heartbeat, so the work happens on schedule and shows up on a board.

1. Start from the role

   The AI software engineer, with these rules in its brief.

2. Put the work on its board

   Add the job as a recurring issue. The employee wakes on its heartbeat, does it and reports in a comment.

3. Related

   The Sentry error triage page has the full brief and priority rules.

FAQ