Atomicwork MCP Server
Enables AI assistants like Claude to interact with Atomicwork tickets, requests, and workflows through natural language, providing tools for listing, searching, and managing tickets.
README
Atomicwork MCP Server
A Model Context Protocol (MCP) server that integrates Atomicwork with Claude Desktop and other MCP clients. This allows AI assistants like Claude to interact with your Atomicwork tickets, requests, and workflows directly.
What is Atomicwork?
Atomicwork is an enterprise service management platform that helps IT teams manage tickets, requests, changes, and incidents. This MCP server brings that functionality into your AI workflow.
Features
- š List Tickets - View all tickets assigned to you
- š Search Tickets - Find tickets by keyword or criteria
- š Get Ticket Details - Retrieve complete information about any ticket
- š Workspace Access - List all requests in your workspace
- š¤ Filter by User - List requests by requester email
- š¬ View Comments - Get all comments and notes on tickets
Requirements
- Node.js 18.0 or higher
- Atomicwork Account with API access
- Claude Desktop or another MCP-compatible client
Installation
Option 1: Clone and Build from Source
# Clone the repository
git clone https://github.com/keller-a16z/public-atomicwork-mcp.git
cd public-atomicwork-mcp
# Install dependencies
npm install # or: bun install
# Build the project
npm run build # or: bun run build
Configuration
Step 1: Get Your Atomicwork API Key
- Log into your Atomicwork instance
- Navigate to Settings ā API (usually at
https://your-company.atomicwork.com/settings/api) - Generate a new API key
- Copy the key - you'll need it for configuration
Step 2: Find Your User ID and Workspace ID
Finding your User ID:
- Open your browser's Developer Tools (F12)
- Go to the Network tab
- Navigate to any page in Atomicwork
- Look for API requests to
/api/v1/users/meor similar - Your user ID will be in the response (usually a number like
12345)
Finding your Workspace ID:
- In the same Network tab
- Look for API requests to
/api/v1/workspaces/ - Your workspace ID will be in the URL (usually a number like
123or456)
Step 3: Configure Claude Desktop
Add the MCP server to your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"atomicwork": {
"command": "node",
"args": [
"/ABSOLUTE/PATH/TO/public-atomicwork-mcp/dist/index.js"
],
"env": {
"ATOMICWORK_API_KEY": "your_api_key_here",
"ATOMICWORK_BASE_URL": "https://your-company.atomicwork.com/api/v1",
"ATOMICWORK_USER_ID": "your_user_id",
"ATOMICWORK_WORKSPACE_ID": "your_workspace_id"
}
}
}
}
Replace:
/ABSOLUTE/PATH/TO/public-atomicwork-mcp/dist/index.jswith the actual path to the built serveryour_api_key_herewith your Atomicwork API keyyour-companywith your Atomicwork instance subdomainyour_user_idwith your user IDyour_workspace_idwith your workspace ID
Step 4: Restart Claude Desktop
Close and reopen Claude Desktop to load the new MCP server.
Usage
Once configured, you can interact with Atomicwork through natural language in Claude Desktop:
Example Queries
List your tickets:
Show me my current Atomicwork tickets
Get ticket details:
Get details for Atomicwork ticket ITREQ-1234
Search for tickets:
Search for tickets about "password reset"
List all workspace requests:
Show me all open requests in the workspace
Filter requests by user:
Show me all tickets from user@company.com
List requests created by john@company.com
View ticket comments:
Show me comments on ticket ITREQ-1234
Get all notes for this ticket
Available Tools
list_my_tickets
Lists all tickets assigned to you.
Parameters:
status(optional): Filter by status (open,in_progress,resolved,closed)limit(optional): Maximum number of tickets to return (default: 50)
Example:
List my in-progress tickets
get_ticket_details
Gets detailed information about a specific ticket.
Parameters:
ticket_id(required): The ID of the ticket
Example:
Show me ticket ITREQ-1234
search_tickets
Searches for tickets by keyword.
Parameters:
query(required): Search query stringassigned_to_me(optional): Only show tickets assigned to you (default: true)
Example:
Find tickets mentioning "GitHub"
list_all_requests
Lists all requests in the workspace (not just yours). Can filter by requester email.
Parameters:
status(optional): Filter by statuslimit(optional): Maximum number of requests (default: 50)requester_email(optional): Filter by requester email address (e.g.,user@company.com)
Examples:
Show all open requests in the workspace
Show me all tickets from john@company.com
List requests from jane@company.com
get_ticket_comments
Gets all comments and notes on a specific ticket.
Parameters:
ticket_id(required): The ID or display_id of the ticket (e.g.,ITREQ-1234)
Example:
Show me comments on ticket ITREQ-1234
Get all notes for ticket 567890
Troubleshooting
MCP Server Not Appearing
- Check the configuration file path - Ensure you edited the correct
claude_desktop_config.json - Verify the dist path - Make sure the path to
dist/index.jsis absolute and correct - Restart Claude Desktop - Changes require a full restart
- Check logs - Look for errors in Claude Desktop's logs
API Connection Errors
"Failed to connect to Atomicwork"
- Verify your API key is correct
- Check that your base URL matches your Atomicwork instance
- Ensure you have network access to your Atomicwork instance
- Try accessing
https://your-company.atomicwork.com/api/v1/users/mein your browser with the API key
"Access Token X-Api-Key is missing/invalid"
- Double-check your API key in the configuration
- Ensure the API key hasn't expired
- Try regenerating a new API key in Atomicwork
No Tickets Returned
"Error: ATOMICWORK_USER_ID not set"
- Make sure you've added your user ID to the configuration
- Verify the user ID is correct (it should be a number)
"Error: ATOMICWORK_WORKSPACE_ID not set"
- Add your workspace ID to the configuration
- Verify the workspace ID is correct (it should be a number)
Build Errors
If you're building from source and encounter errors:
# Clean and rebuild
rm -rf node_modules dist
npm install # or: bun install
npm run build # or: bun run build
Development
Running Locally
# Install dependencies
bun install
# Build the project
bun run build
# Test the MCP server
node dist/index.js
Project Structure
public-atomicwork-mcp/
āāā src/
ā āāā index.ts # Main MCP server implementation
āāā dist/ # Compiled JavaScript output
āāā package.json # Project dependencies and scripts
āāā tsconfig.json # TypeScript configuration
āāā .env.example # Environment variable template
āāā .gitignore # Git ignore rules
āāā README.md # This file
Security
- Never commit your API keys - Use environment variables
- Keep your API key secure - Don't share it publicly
- Regenerate compromised keys - If your key is exposed, regenerate it immediately in Atomicwork
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
License
MIT License - see LICENSE file for details
Links
Support
For issues and questions:
- Bug reports: GitHub Issues
- Discussions: GitHub Discussions
Acknowledgments
Inspired by the ClickUp MCP Server by taazkareem.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.