Sentry Error Triage

What you can do

A few things this template does well out of the box.

• Summarise the top 5 new errors from the last 24 hours with suggested owners.
• Walk through a specific stack trace and explain what broke.
• Find which release introduced a particular issue.
• Suggest a fix based on the error context and recent commits.

How it works

Three steps to go from template to a live chat.

MCP servers used

The endpoints this template connects to by default. You can swap any of them in Agent Studio.

Getting your access token

A quick walkthrough for the credential this template needs.

1. 1In Sentry, go to Settings → Account → Auth Tokens.
2. 2Create a token with `project:read`, `event:read`, `org:read` scopes.
3. 3Copy the token (starts with `sntrys_`) and paste it into Agent Studio.

Try these prompts

Copy one into the studio to see the agent in action.

• What new errors appeared in production after the last deploy?

• Summarise issue ABC-1234 and tell me the likely root cause.

• Find all errors involving `TypeError` in the checkout flow this week.

System prompt

The default instructions the model starts with. Edit it any time inside Agent Studio.

You are a production error triage assistant connected to Sentry.

Use the available tools to:
- List and inspect recent issues, including stack traces and breadcrumbs
- Group related errors and identify the root cause
- Find the release or commit that introduced an issue
- Suggest fixes or follow-up actions (open a ticket, assign to a team)

Be concise. Structure your answer as: what broke, where, likely cause, suggested next step.