Cursor Admin
About
Monitor team productivity, usage, and spending for AI assistants using the Cursor Admin API.
Details
- Author
- h3ro-dev
- Categories
- Productivity, Other, Infrastructure, AI
Jump to
Setup
Install Cursor Admin in your MCP client (Claude Desktop, Cursor, Windsurf, and others).
Repository: https://github.com/h3ro-dev/cursor-admin-mcp
Follow the installation instructions in the repository README, then restart your MCP client.
๐ฎ The Team Analytics Power-Up for Your AI Assistant
Brought to you byUtlyze- Zero Latency Operations Through AI Employees
That's it. Your AI assistant just leveled up.
Think of this as equipping your AI with X-ray vision for your development team:
- ๐ข Team Roster- Instantly know who's on the team and their roles
- ๐ Usage Analytics- See who's actually coding vs. who's in meetings
- ๐ฐ Credit Tracker- Know exactly where your AI credits are going
- ๐ Productivity Metrics- Lines added, AI acceptance rates, favorite models
- ๐ Deep Insights- Which developer uses which tools, when they're most active
Stop asking your team for status updates. Your AI can now tell you:
- "Who wrote the most code this week?"
- "What's our AI spending trend?"
- "Which team member has the highest AI acceptance rate?"
- "Show me usage patterns for the last month"
You: "Show me team usage for this week" AI: Here's your team's activity for the last 7 days: ๐ฅ TEAM MEMBERS 1. James Brady (Owner) - james@jamesbrady.org 2. Cody Vincent - vincent.cody298@gmail.com 3. 2 other members ๐ USAGE STATISTICS โข Most Active: Cody Vincent (641 AI requests) โข Most Productive: Cody Vincent (29,477 lines added) โข Top Models: claude-4-opus, gpt-4, claude-sonnet โข Active Days: 2/4 members coding daily ๐ฐ SPENDING: $1,957.32 total โข James Brady: $1,848.57 (94.4%) โข Cody Vincent: $73.76 (3.8%) โข Others: $34.99 (1.8%)
- Node.js 16+
- A Cursor team admin API key (get it from your team settings)
git clone https://github.com/h3ro-dev/cursor-admin-mcp.git cd cursor-admin-mcp npm install npm run build
Create a.envfile in the project root (or set the environment variable):
CURSOR_API_KEY=key_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):
{ "mcpServers": { "cursor-admin": { "command": "npx", "args": ["cursor-admin-mcp"], "env": { "CURSOR_API_KEY": "key_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" } } } }
{ "mcpServers": { "cursor-admin": { "command": "node", "args": ["/path/to/cursor-admin-mcp/dist/index.js"], "env": { "CURSOR_API_KEY": "key_xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx" } } } }
Get a list of all team members with their information.
"Show me all team members" "List everyone on the team with their roles"
[ { "name": "John Doe", "email": "john@example.com", "role": "admin" }, { "name": "Jane Smith", "email": "jane@example.com", "role": "member" } ]
Retrieve detailed usage metrics for a specified date range (max 90 days).
- startDate: Start date in epoch milliseconds
- endDate: End date in epoch milliseconds
"Show me usage data for the last 7 days" "What was our team's AI acceptance rate last month?" "Which models did we use most this week?"
[ { "date": "2024-01-15", "linesAdded": 1523, "linesDeleted": 342, "acceptanceRate": 0.82, "requestTypes": { "completion": 234, "chat": 56, "edit": 23 }, "mostUsedModels": ["gpt-4", "claude-3"], "mostUsedExtensions": ["copilot", "cursor-tab"], "clientVersion": "0.42.0" } ]
Get team spending information with optional filtering and pagination.
- searchTerm: Filter by search term
- sortBy: Field to sort by
- sortDirection: "asc" or "desc"
- page: Page number
- pageSize: Items per page
"Show me team spending" "Who are the top spenders this month?" "Search for John's spending data"
{ "members": [ { "email": "john@example.com", "name": "John Doe", "spending": 125.50 } ], "total": 1250.00, "page": 1, "pageSize": 10 }
# Clone the repository git clone https://github.com/h3ro-dev/cursor-admin-mcp.git cd cursor-admin-mcp # Install dependencies npm install # Copy environment file cp .env.example .env # Edit .env and add your CURSOR_API_KEY # Run in development mode npm run dev
# Run tests npm test # Run tests with coverage npm test -- --coverage # Run tests in watch mode npm run test:watch
# Build the project npm run build # Run the built version npm start
cursor-admin-mcp/ โโโ src/ โ โโโ index.ts # MCP server implementation โ โโโ cursor-client.ts # Cursor API client โโโ tests/ โ โโโ cursor-client.test.ts # Test suite โโโ examples/ โ โโโ (example scripts) โโโ docs/ โ โโโ (additional documentation) โโโ package.json โโโ tsconfig.json โโโ jest.config.js โโโ .env.example โโโ .gitignore โโโ README.md
// The MCP server handles all the communication // Just use natural language in Claude/Cursor: "Show me all team members and their roles" "Get usage data for January 2024" "What's our team's total spending this month?" "Show me the AI acceptance rate trend for the last 30 days"
// Complex date ranges "Compare usage between last week and this week" // Specific metrics "Which team members have the highest AI suggestion acceptance rate?" // Spending analysis "Show me spending sorted by amount for team members whose name contains 'John'"
The server includes comprehensive error handling:
- Invalid API Key: Clear error message if authentication fails
- Rate Limiting: Respects Cursor API rate limits
- Date Validation: Ensures date ranges don't exceed 90 days
- Network Errors: Graceful handling of connection issues
- Invalid Parameters: Detailed validation messages
- API keys are never logged or exposed
- All communication uses HTTPS
- Basic authentication as per Cursor API requirements
- Environment variables for sensitive data
- Fork the repository
- Create a feature branch (git checkout -b feature/amazing-feature)
- Commit your changes (git commit -m 'Add amazing feature')
- Push to the branch (git push origin feature/amazing-feature)
- Open a Pull Request
- Write tests for new features
- Update documentation
- Follow TypeScript best practices
- Ensure all tests pass
- Add examples for new functionality
- Verify your API key starts withkey_
- Ensure you're a team admin
- Check the key hasn't been revoked
- Check your internet connection
- Verify the Cursor API is accessible
- Check for any firewall restrictions
- Ensure your date range is within 90 days
- Use epoch milliseconds for dates
- Verify endDate > startDate
MIT License - see LICENSE file for details
- Built for theCursoreditor
- Implements theModel Context Protocol
- Uses the officialMCP SDK
- ๐Report bugs
- ๐กRequest features
- ๐Read the docs
- ๐ฌJoin discussions
Utlyze - Zero Latency Operations Through AI Employees
We're pioneering the future of work by creating AI employees that seamlessly integrate with your existing workflows. The Cursor Admin MCP Server is just one example of how we're making teams more efficient through intelligent automation.
Visit Utlyze.com|Contact Us|LinkedIn
Made with โค๏ธ by the Utlyze team for the Cursor community
AI-powered wellbeing insights with real-time benchmarks and alerts.
Provides structured access to ActivityWatch time tracking data for AI assistants.
Access fathom ai api endpoints via mcp server and llm.
Multi-Agent Monitoring LangFuse MCP Server
A Model Context Protocol (MCP) server for comprehensive monitoring and observability of multi-agent systems using Langfuse.
Predict and track AI ROI using Monte Carlo simulations, real-time industry benchmarks, and ML-powered insights.
Stubble learns how you work and organizes your day automatically. Export timesheets, surface insights, and supercharge your AI tools.
Paid remote MCP for AI agent run monitoring, failure detection, tool-call incident replay, SLA receipts, and client status exports.
Behavioral trust layer for the AI agent economy. Check MCP server reliability scores, report interactions, detect anomalies, and discover the most trusted servers. 8 tools, free, real-time.
Behavioral trust scoring for MCP servers and AI agents. Live registry tracking 4,500+ servers with trust scores based on interaction history, success rates, and latency.
Behavioral trust scoring for 14,820+ MCP servers. Check reliability, latency, and success rates before tool calls.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.




