Sentry Issue Analyzer MCP Server

MonitoringPython

Retrieve and analyze issues from Sentry.io

Available Tools

get_sentry_issue

Retrieve and analyze a Sentry issue by ID or URL, providing details including title, issue ID, status, level, timestamps, event count, and full stacktrace

issue_id_or_url

The Sentry Issue Analyzer MCP provides a seamless interface to access and analyze error reports, stacktraces, and debugging information from your Sentry.io account. This integration allows you to quickly retrieve detailed information about specific issues without leaving your development environment. With this MCP, you can inspect error details including titles, status, severity levels, timestamps, event counts, and full stacktraces. This makes it an invaluable tool for developers looking to troubleshoot application errors efficiently while working with AI assistants.

Overview

The Sentry Issue Analyzer MCP server connects to your Sentry.io account to retrieve and analyze error reports. This integration allows you to quickly access detailed information about specific issues directly within your AI assistant workflow.

Installation

There are several ways to install and run the Sentry Issue Analyzer MCP:

Using uv (Recommended)

If you have uv installed, you can run the MCP directly without a separate installation step using uvx:

uvx mcp-server-sentry --auth-token YOUR_SENTRY_TOKEN

Using pip

You can install the MCP via pip:

pip install mcp-server-sentry

After installation, run it as a Python module:

python -m mcp_server_sentry --auth-token YOUR_SENTRY_TOKEN

Using Docker

You can also run the MCP using Docker:

docker run -i --rm mcp/sentry --auth-token YOUR_SENTRY_TOKEN

Configuration

Authentication

You'll need a Sentry authentication token to use this MCP. You can obtain this from your Sentry account settings.

The token can be provided in two ways:

As a command-line argument: --auth-token YOUR_SENTRY_TOKEN
As an environment variable: SENTRY_AUTH_TOKEN=YOUR_SENTRY_TOKEN

Client Integration

Claude Desktop

Add this configuration to your claude_desktop_config.json:

"mcpServers": {
  "sentry": {
    "command": "uvx",
    "args": ["mcp-server-sentry", "--auth-token", "YOUR_SENTRY_TOKEN"]
  }
}

VS Code

For VS Code, you can add the following to your settings:

{
  "mcp": {
    "inputs": [
      {
        "type": "promptString",
        "id": "auth_token",
        "description": "Sentry Auth Token",
        "password": true
      }
    ],
    "servers": {
      "sentry": {
        "command": "uvx",
        "args": ["mcp-server-sentry"],
        "env": {
          "SENTRY_AUTH_TOKEN": "${input:auth_token}"
        }
      }
    }
  }
}

Zed

For Zed, add to your settings.json:

"context_servers": {
  "mcp-server-sentry": {
    "command": {
      "path": "uvx",
      "args": ["mcp-server-sentry", "--auth-token", "YOUR_SENTRY_TOKEN"]
    }
  }
},

Usage

Once configured, you can use the MCP to retrieve and analyze Sentry issues. Simply provide an issue ID or URL to get detailed information about the error.

Debugging

To debug the MCP server, you can use the MCP inspector:

npx @modelcontextprotocol/inspector uvx mcp-server-sentry --auth-token YOUR_SENTRY_TOKEN

License

This MCP server is licensed under the MIT License.