Skip to main content
Connect Sentry to Tembo for automatic error detection and code fixes. When Sentry detects an error in your application, Tembo analyzes the stack trace and creates a pull request with a targeted fix.

​
Features

  • Automatic error monitoring - Detects new errors as they occur in your production environment
  • Stack trace analysis - Analyzes error context, breadcrumbs, and related data to understand root causes
  • Intelligent fixes - Generates pull requests with code changes that address the specific error
  • Webhook-driven - Responds immediately when Sentry reports new issues

​
Installation

1

Connect

Navigate to the Integrations page in Tembo and click the Connect button next to Sentry.
2

Authorize

Authorize Tembo to access your Sentry account with permissions to read error data and receive webhook events. You’ll be redirected back to the Integrations page when authorization is complete.
3

Map Projects

Map your Sentry projects to GitHub repositories under “Projects” on the Integrations page.Only mapped projects will generate pull requests with code fixes.

​
Supported Webhooks

Tembo listens to Sentry issue lifecycle events and maps them to automation triggers (reference them as sentry.<event> in your triggerName). The Integrations page in the app shows the live set enabled for your org. Supported events:
  • issue.created
  • issue.regression
  • issue.resolved
  • issue.ignored
  • issue.assigned
See Automations for creating triggers from these events.

​
How It Works

  1. Error Detected - Sentry captures an error in your application and sends a webhook to Tembo
  2. Analysis - Tembo retrieves the full error context, including stack traces, breadcrumbs, and environment details
  3. Solution Generation - Tembo analyzes your codebase and creates an appropriate fix
  4. Pull Request - A PR is automatically opened in your repository with the fix and context
  5. Review & Iterate - Use the Feedback Loop to refine the solution if needed

​
Best Practices

​
Project Mapping

  • Map Sentry projects to the repositories where the errors actually occur
  • For monorepos, you may need multiple Sentry projects mapped to the same repository
  • Verify mappings are correct to ensure fixes target the right codebase

​
Error Context

To help Tembo generate better fixes:
  • Configure Sentry to capture comprehensive stack traces
  • Enable source maps for minified JavaScript/TypeScript
  • Include relevant tags and custom context in your Sentry configuration
  • Set up breadcrumbs to track user actions leading to errors

​
Reviewing Auto-Generated Fixes

  • Always review PRs before merging - Tembo provides solutions, you decide what gets deployed
  • Use the Feedback Loop to request improvements by tagging @tembo in PR comments
  • Test the fix in a staging environment before deploying to production

​
MCP Server Integration

Tembo provides a Model Context Protocol (MCP) server for Sentry, enabling AI coding agents to directly interact with your Sentry data during task execution.

​
What is MCP?

Model Context Protocol is a standardized way for AI applications to connect to external data sources and tools. With MCP, agents can query Sentry error data, retrieve stack traces, and access issue details in real-time while solving problems.

​
Using the Sentry MCP Server

When you connect Sentry to Tembo, the MCP server is automatically available to agents working on your tasks. Agents can:
  • Query error data - Search and filter Sentry issues programmatically
  • Retrieve stack traces - Access detailed error information for debugging
  • Check issue status - Verify if errors are resolved or still occurring
  • Analyze error trends - Understand error patterns and frequency
This enables agents to make more informed decisions when fixing bugs, as they have direct access to production error data and can verify their fixes against actual error conditions.