Bugsnag MCP Server

A Model Context Protocol (MCP) server for interacting with Bugsnag. This server allows LLM tools like Cursor and Claude to investigate and resolve issues in Bugsnag.

✨ Features

Error Monitoring & Analysis

• Organization & Project Navigation: Easily browse your Bugsnag hierarchy
• Error & Event Filtering: Find specific issues with powerful filtering options
• Detailed Stacktrace Viewing: See formatted stacktraces with source code context and highlighted error lines
• Exception Chain Visualization: Understand the root cause by viewing the full exception chain

Code Intelligence

• Project vs. Library Code Distinction: Clearly identify your code vs third-party libraries
• Source Code Context: View relevant code snippets around error locations
• Error Patterns: Identify recurring patterns across multiple errors

Issue Management

• Search Capabilities: Find issues by error class, message, or app version
• Error Details: Get comprehensive information about each error
• Event History: View all occurrences of a specific error

🚀 Quick Setup

Setting up the Bugsnag MCP server is simple and doesn't require any installation or downloading source code.

For Cursor

1. Add the Bugsnag MCP server configuration:

   config.json

   Copy

   {
     "mcpServers": {
       "bugsnag": {
         "command": "npx",
         "args": ["-y", "bugsnag-mcp-server"],
         "env": {
           "BUGSNAG_API_KEY": "your-bugsnag-api-key"
         },
         "disabled": false,
         "alwaysAllow": []
       }
     }
   }
   

2. Replace your-bugsnag-api-key with your Bugsnag API key

For Claude Desktop

1. Add the Bugsnag MCP server configuration:

   config.json

   Copy

   {
     "mcpServers": {
       "bugsnag": {
         "command": "npx",
         "args": ["-y", "bugsnag-mcp-server"],
         "env": {
           "BUGSNAG_API_KEY": "your-bugsnag-api-key"
         },
         "disabled": false,
         "alwaysAllow": []
       }
     }
   }
   

2. Replace your-bugsnag-api-key with your Bugsnag API key

🔑 Obtaining a Bugsnag API Key

To use this MCP server, you'll need a Bugsnag API key:

1. Log in to your Bugsnag account at https://app.bugsnag.com/
2. Go to Settings > Organization settings > Access tokens
3. Create a new personal access token with the following permissions:
   • Read projects
   • Read and write errors
   • Read and write comments
4. Copy the generated token for use with the MCP server

📋 Usage Examples

Once configured, you can use the Bugsnag MCP server with your LLM tool. Here are some example prompts:

Exploring Your Bugsnag Account

List all my Bugsnag organizations

Show me all projects in organization "org_12345"

Finding and Analyzing Errors

List the open errors in my Bugsnag project "project_12345"

Show me the details for Bugsnag error ID "error_12345"

Show me the detailed stacktrace for event "event_12345" in project "project_12345"

View the exception chain for event "event_12345" in project "project_12345"

Searching for Specific Issues

Search for Bugsnag issues in project "project_12345" related to "NullPointerException"

List all events for error "error_12345" in project "project_12345"

📚 Available Tools

The Bugsnag MCP server provides the following tools:

Organization & Project Management

list_organizations

Lists available Bugsnag organizations.

Parameters:

• None required

list_projects

Lists projects in an organization.

Parameters:

• organization_id (required): Bugsnag organization ID

Error & Event Management

list_errors

Lists errors in a project with filtering options.

Parameters:

• project_id (required): Bugsnag project ID
• status: Filter by error status ("open", "fixed", "ignored")
• sort: Sort order for errors ("newest", "oldest", "priority")
• limit: Maximum number of errors to return

view_error

Gets detailed information about a specific error.

Parameters:

• error_id (required): Bugsnag error ID

list_error_events

Lists events (occurrences) for a specific error.

Parameters:

• project_id (required): Bugsnag project ID
• error_id (required): Bugsnag error ID
• limit: Maximum number of events to return

view_latest_event

Views the latest event for an error.

Parameters:

• error_id (required): Bugsnag error ID

view_event

Views detailed information about a specific event.

Parameters:

• project_id (required): Bugsnag project ID
• event_id (required): Bugsnag event ID

Stacktrace Analysis

view_stacktrace

Extracts and formats stacktrace information from an event.

Parameters:

• project_id (required): Bugsnag project ID
• event_id (required): Bugsnag event ID
• include_code: Include source code context if available (default: true)

view_exception_chain

Views the full chain of exceptions for an event.

Parameters:

• project_id (required): Bugsnag project ID
• event_id (required): Bugsnag event ID

Issue Management

search_issues

Searches for issues using various criteria.

Parameters:

• project_id (required): Bugsnag project ID
• query: Search query
• error_class: Filter by error class
• app_version: Filter by app version

🛠️ Advanced Usage

Testing Your API Key

You can test if your Bugsnag API key is valid using:

npx bugsnag-mcp-server test-api-key your-bugsnag-api-key

If your API key is valid, this will display a list of your Bugsnag projects with their IDs.

Installation Options

While using npx is recommended for most users, you can also install the package globally if you prefer:

npm install -g bugsnag-mcp-server

For Developers

If you're interested in contributing or modifying the code:

1. Clone the repository:

   bash

   Copy

   git clone https://github.com/yourusername/bugsnag-mcp.git
   cd bugsnag-mcp
   

2. Install dependencies:

   Terminal

   Copy

   npm install
   

3. Build the project:

   Terminal

   Copy

   npm run build
   

4. Run tests:

   Terminal

   Copy

   npm test
   

5. Run tests with coverage:

   Terminal

   Copy

   npm run test:coverage
   

6. Format code:

npm run format

7. Check code formatting:

npm run format:check

Continuous Integration

This project uses GitHub Actions for continuous integration. The workflows automatically:

• Runs on push to the main branch and on pull requests
• Tests against multiple Node.js versions (18.x and 20.x)
• Runs the test suite
• Generates and uploads test coverage reports
• Checks code formatting with Prettier

You can view the workflow configurations in:

• .github/workflows/test.yml - For running tests
• .github/workflows/prettier.yml - For checking code formatting

License

MIT