Cloud Agent MCP Server
Enables AI assistants to create and manage Cursor Cloud Agents that autonomously work on GitHub repositories, including creating tasks, monitoring progress, and automatically generating pull requests.
README
Cloud Agent MCP Server
MCP server for the Cursor Cloud Agents API. Lets AI assistants create and manage cloud agents that work on GitHub repositories.
Quick Start
# Install
npm install -g cursor-cloud-agent-mcp
# Set your API key
export CURSOR_API_KEY=your_api_key_here
# Use with Cursor (create .cursor/mcp.json)
{
"mcpServers": {
"cursor-cloud-agent": {
"command": "npx",
"args": ["-y", "cursor-cloud-agent-mcp"],
"env": {
"CURSOR_API_KEY": "${env:CURSOR_API_KEY}"
}
}
}
}
Installation
Install from npm
npm install -g cursor-cloud-agent-mcp
Or install locally in your project:
npm install cursor-cloud-agent-mcp
Install from Source
If you're developing or want to run from source:
# Clone and install dependencies
git clone https://github.com/jxnl/cursor-cloud-agent-mcp
cd cloud-agent-mcp
npm install
# Set your API key
export CURSOR_API_KEY=your_api_key_here
# Run HTTP server
npm start
# Server runs at http://localhost:3000/mcp
# Or run stdio server
npm run start:stdio
Configuration
Environment Variables
| Variable | Required | Description |
|---|---|---|
CURSOR_API_KEY |
Yes | API key from cursor.com/settings |
PORT |
No | Server port for HTTP version only (default: 3000) |
Connecting Clients
Cursor
Option 1: Using npm package (Recommended)
After installing via npm, create .cursor/mcp.json:
{
"mcpServers": {
"cursor-cloud-agent": {
"command": "npx",
"args": ["-y", "cursor-cloud-agent-mcp"],
"env": {
"CURSOR_API_KEY": "${env:CURSOR_API_KEY}"
}
}
}
}
Or if installed globally:
{
"mcpServers": {
"cursor-cloud-agent": {
"command": "cursor-cloud-agent-mcp",
"env": {
"CURSOR_API_KEY": "${env:CURSOR_API_KEY}"
}
}
}
}
Option 2: From source (Development)
If running from source, create .cursor/mcp.json:
{
"mcpServers": {
"cursor-cloud-agent": {
"command": "npm",
"args": ["run", "start:stdio"],
"env": {
"CURSOR_API_KEY": "${env:CURSOR_API_KEY}"
}
}
}
}
Option 3: HTTP Server (Alternative)
If you prefer the HTTP version, configure it as:
{
"mcpServers": {
"cursor-cloud-agent": {
"url": "http://localhost:3000/mcp",
"headers": {}
}
}
}
Then run npm start in a separate terminal to start the HTTP server.
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"cloud-agent": {
"type": "http",
"url": "http://localhost:3000/mcp"
}
}
}
MCP Inspector
npx @modelcontextprotocol/inspector
# Connect to http://localhost:3000/mcp
Usage
Typical Workflow
1. get_repos → Get current repo URL and branch
2. create_task → Launch task with prompt
3. get_task → Check status (CREATING → RUNNING → FINISHED)
4. add_followup → (optional) Send more instructions while running
5. get_conversation → Review what the task did
Available Tools
Discovery Tools
| Tool | Description |
|---|---|
get_repos |
Detects current git repo (repository URL, branch, uncommitted changes). Optionally lists all accessible repos. Call this first. Note: When using include_all: true, filters (regex_patterns) are REQUIRED. Fetching all repos has strict rate limits (1/min, 30/hour). |
get_me |
Get API key info (verify authentication) |
get_models |
List available LLM models |
Task Lifecycle Tools
| Tool | Description |
|---|---|
create_task |
Launch a cloud task with a task prompt |
list_tasks |
List all your cloud tasks |
get_task |
Get status of a specific task |
add_followup |
Send additional instructions to a running task |
get_conversation |
Get full conversation history |
delete_task |
Permanently delete a task |
Common Examples
Create a Task
{
"tool": "create_task",
"arguments": {
"prompt": "Add a README.md file with installation instructions",
"repository": "https://github.com/your-org/your-repo",
"auto_pr": true
}
}
List Tasks
{
"tool": "list_tasks",
"arguments": {
"filter": "FINISHED|RUNNING",
"limit": 10
}
}
Get Repository Info
{
"tool": "get_repos"
}
Reference
Tool Documentation
get_repos - Get Repositories
Basic usage (current repo only):
{
"tool": "get_repos"
}
Fetch all accessible repositories (filters REQUIRED):
Important: When using include_all: true, you MUST provide regex_patterns to filter the results. This prevents returning too many repositories.
Filter repositories with regex patterns:
{
"tool": "get_repos",
"arguments": {
"include_all": true,
"regex_patterns": ["^my-.*"]
}
}
Multiple filter patterns (OR logic):
{
"tool": "get_repos",
"arguments": {
"include_all": true,
"regex_patterns": [".*api.*", ".*backend.*", ".*frontend.*"]
}
}
Specify working directory:
{
"tool": "get_repos",
"arguments": {
"working_directory": "/path/to/project"
}
}
Filter Examples:
["^my-.*"]- Repos starting with "my-"[".*api.*"]- Repos containing "api"["github.com/myorg"]- Repos from specific org[".*backend.*", ".*server.*"]- Repos matching either pattern
get_me - Verify Authentication
{
"tool": "get_me"
}
get_models - List Available Models
{
"tool": "get_models"
}
create_task - Launch a Cloud Task
Basic usage:
{
"tool": "create_task",
"arguments": {
"prompt": "Add a README.md file with installation instructions",
"repository": "https://github.com/your-org/your-repo"
}
}
With branch specification:
{
"tool": "create_task",
"arguments": {
"prompt": "Fix authentication bug",
"repository": "https://github.com/your-org/your-repo",
"ref": "main"
}
}
Auto-create PR:
{
"tool": "create_task",
"arguments": {
"prompt": "Add user authentication",
"repository": "https://github.com/your-org/your-repo",
"auto_pr": true
}
}
Custom branch name:
{
"tool": "create_task",
"arguments": {
"prompt": "Implement new feature",
"repository": "https://github.com/your-org/your-repo",
"branch_name": "feature/new-feature"
}
}
Specify model:
{
"tool": "create_task",
"arguments": {
"prompt": "Refactor codebase",
"repository": "https://github.com/your-org/your-repo",
"model": "claude-4-sonnet-thinking"
}
}
With plan file:
{
"tool": "create_task",
"arguments": {
"prompt": "Implement the features described in the plan",
"repository": "https://github.com/your-org/your-repo",
"plan_file": "./plan.md"
}
}
Complete example with all options:
{
"tool": "create_task",
"arguments": {
"prompt": "Add comprehensive test coverage",
"repository": "https://github.com/your-org/your-repo",
"ref": "develop",
"branch_name": "feature/add-tests",
"auto_pr": true,
"model": "o3",
"plan_file": "./test-plan.md"
}
}
list_tasks - List All Tasks
Basic listing:
{
"tool": "list_tasks"
}
Limit results:
{
"tool": "list_tasks",
"arguments": {
"limit": 10
}
}
Pagination with cursor:
{
"tool": "list_tasks",
"arguments": {
"limit": 20,
"cursor": "bc_ghi789"
}
}
Filter by status:
{
"tool": "list_tasks",
"arguments": {
"filter": "FINISHED|RUNNING"
}
}
Filter by repository:
{
"tool": "list_tasks",
"arguments": {
"filter": ".*my-repo.*"
}
}
Filter by branch name:
{
"tool": "list_tasks",
"arguments": {
"filter": "feature/.*"
}
}
Filter by summary content:
{
"tool": "list_tasks",
"arguments": {
"filter": ".*README.*"
}
}
Combine filters:
{
"tool": "list_tasks",
"arguments": {
"filter": "FINISHED.*my-repo"
}
}
Filter Examples:
"FINISHED"- Only finished tasks"RUNNING|CREATING"- Active tasks".*api.*"- Tasks working on API repos"feature/.*"- Tasks on feature branches"FINISHED.*README"- Finished tasks with README in summary"FAILED|CANCELLED"- Failed or cancelled tasks
get_task - Get Task Status
{
"tool": "get_task",
"arguments": {
"id": "bc_abc123"
}
}
add_followup - Send Follow-up Instructions
{
"tool": "add_followup",
"arguments": {
"id": "bc_abc123",
"prompt": "Also add a troubleshooting section"
}
}
get_conversation - Get Conversation History
{
"tool": "get_conversation",
"arguments": {
"id": "bc_abc123"
}
}
delete_task - Delete a Task
{
"tool": "delete_task",
"arguments": {
"id": "bc_abc123"
}
}
Response Shapes
All tools return structured JSON responses matching the Cloud Agents API specification. See docs.md for complete API documentation.
get_me Response
{
"apiKeyName": "Production API Key",
"createdAt": "2024-01-15T10:30:00Z",
"userEmail": "developer@example.com"
}
get_models Response
{
"models": [
"claude-4-sonnet-thinking",
"o3",
"claude-4-opus-thinking"
]
}
get_repos Response
{
"current": {
"repository": "https://github.com/your-org/your-repo",
"branch": "main",
"has_uncommitted_changes": false
},
"available": [
{
"owner": "your-org",
"name": "your-repo",
"repository": "https://github.com/your-org/your-repo"
}
],
"total_count": 1
}
create_task Response
{
"id": "bc_abc123",
"name": "Add README Documentation",
"status": "CREATING",
"source": {
"repository": "https://github.com/your-org/your-repo",
"ref": "main"
},
"target": {
"branchName": "feature/add-readme",
"url": "https://cursor.com/agents?id=bc_abc123",
"autoCreatePr": true,
"openAsCursorGithubApp": false,
"skipReviewerRequest": false
},
"createdAt": "2024-01-15T10:30:00Z"
}
list_tasks Response
{
"tasks": [
{
"id": "bc_abc123",
"name": "Add README Documentation",
"status": "FINISHED",
"source": {
"repository": "https://github.com/your-org/your-repo",
"ref": "main"
},
"target": {
"branchName": "cursor/add-readme-1234",
"url": "https://cursor.com/agents?id=bc_abc123",
"prUrl": "https://github.com/your-org/your-repo/pull/1234",
"autoCreatePr": false,
"openAsCursorGithubApp": false,
"skipReviewerRequest": false
},
"summary": "Added README.md with installation instructions and usage examples",
"createdAt": "2024-01-15T10:30:00Z"
}
],
"nextCursor": "bc_ghi789"
}
get_task Response
{
"id": "bc_abc123",
"name": "Add README Documentation",
"status": "FINISHED",
"source": {
"repository": "https://github.com/your-org/your-repo",
"ref": "main"
},
"target": {
"branchName": "cursor/add-readme-1234",
"url": "https://cursor.com/agents?id=bc_abc123",
"prUrl": "https://github.com/your-org/your-repo/pull/1234",
"autoCreatePr": false,
"openAsCursorGithubApp": false,
"skipReviewerRequest": false
},
"summary": "Added README.md with installation instructions and usage examples",
"createdAt": "2024-01-15T10:30:00Z"
}
Status Values: CREATING, RUNNING, FINISHED, FAILED, CANCELLED
get_conversation Response
{
"id": "bc_abc123",
"messages": [
{
"id": "msg_001",
"type": "user_message",
"text": "Add a README.md file with installation instructions"
},
{
"id": "msg_002",
"type": "assistant_message",
"text": "I'll help you create a comprehensive README.md file..."
}
]
}
add_followup Response
{
"id": "bc_abc123"
}
delete_task Response
{
"id": "bc_abc123"
}
Advanced
Plan Parallel Tasks
Use /plan-parallel-tasks to break down a project for multiple parallel tasks:
/plan-parallel-tasks
project_description: "Add user authentication with login, signup, and password reset"
This will:
- Call
get_reposto detect your repository - Break the project into independent tasks
- Group tasks into phases (parallel → sequential)
- Provide exact prompts for each
create_taskcall
Parallelization Rules
CAN run in parallel:
- Tasks that modify completely different files
- Tasks that create new files without touching shared files
CANNOT run in parallel:
- Tasks that modify the same file
- Tasks where one depends on another's output
- Tasks that both modify package.json, tsconfig.json, etc.
Server Versions
This package includes two server versions:
- HTTP Server (
src/server.ts): Runs an Express HTTP server on port 3000. Use for remote connections or when you need HTTP endpoints. - Stdio Server (
src/server-stdio.ts): Uses standard input/output. Recommended for local integrations. Better for process-based spawning.
Run with:
- HTTP:
npm start(default) - Stdio:
npm run start:stdio
Development
Health Check
The server exposes a health check endpoint:
curl http://localhost:3000/health
Returns: {"status":"ok","service":"cursor-cloud-agent-mcp","version":"1.0.2"}
Publishing to npm
To publish this package to npm:
# Make sure you're logged in
npm login
# Publish
npm publish
The package will be available as cursor-cloud-agent-mcp on npm.
License
MIT
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.
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.
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.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
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.
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.
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.
E2B
Using MCP to run code via e2b.