# 🚀 CrewAI Enterprise MCP Server

A Model Context Protocol (MCP) server that provides access to CrewAI Enterprise API for AI agent orchestration and task execution, deployed on the [Apify platform](https://docs.apify.com/platform/actors).

This Actor enables you to:
- Connect to your CrewAI Enterprise server via MCP
- Start crew tasks with custom inputs
- Monitor crew task status and results
- Monetize your server using Apify's Pay Per Event (PPE) model

## ✨ Features

- **CrewAI Integration**: Direct access to CrewAI Enterprise API endpoints
- **Built-in charging**: Integrated [Pay Per Event (PPE)](https://docs.apify.com/platform/actors/publishing/monetize#pay-per-event-pricing-model) for:
  - Server startup
  - Tool calls (kickoff_crew, get_crew_status)
  - Tool listing
- **Easy configuration**: Simple setup through Actor input or environment variables
- **SSE Transport**: Exposes MCP server via Server-Sent Events for real-time communication

## 🛠 Available Tools

### `kickoff_crew`
Start a new crew task with the provided inputs.

**Parameters:**
- `inputs` (object): Dictionary containing the query and other input parameters for the crew

**Returns:** Dictionary containing the crew task response, including the crew ID needed to check status.

### `get_crew_status`
Get the status of a crew task by its ID.

**Parameters:**
- `crew_id` (string): The ID of the crew task to check

**Returns:** Dictionary containing the crew task status and results.

## 🚀 Quick Start

### 1. Configure the Actor

Set up your CrewAI Enterprise server details either through:

**Actor Input (Recommended):**
- `crewaiServerUrl`: Your CrewAI Enterprise server URL (e.g., `https://your-crewai-server.com/api`)
- `bearerToken`: Bearer token for authenticating with the CrewAI Enterprise API

**Environment Variables:**
- `MCP_CREWAI_ENTERPRISE_SERVER_URL`: CrewAI Enterprise server URL
- `MCP_CREWAI_ENTERPRISE_BEARER_TOKEN`: Bearer token for authentication

### 2. Deploy and Enable Standby Mode

1. Deploy the Actor to Apify
2. Enable standby mode for the Actor
3. Note the Actor's standby URL

### 3. Connect Using an MCP Client

Add the following configuration to your MCP client:

```json
{
    "mcpServers": {
        "crewai-enterprise": {
            "url": "https://your-actor.apify.actor/sse"
        }
    }
}
```

### 4. Use the Tools

Once connected, you can use the CrewAI tools in your MCP client:

```javascript
// Start a crew task
const result = await mcpClient.callTool("kickoff_crew", {
    inputs: {
        query: "Analyze market trends for Q1 2024",
        additional_context: "Focus on technology sector"
    }
});

// Check crew status
const status = await mcpClient.callTool("get_crew_status", {
    crew_id: result.crew_id
});
```

## 💰 Pricing

This Actor uses the [Pay Per Event (PPE)](https://docs.apify.com/platform/actors/publishing/monetize#pay-per-event-pricing-model) monetization model:

- **Server startup**: $0.01 per startup
- **Tool calls**: $0.05 per tool execution (kickoff_crew, get_crew_status)
- **Tool listing**: $0.001 per list operation

## 🔧 Configuration

### Required Configuration

You must provide either through Actor input or environment variables:

1. **CrewAI Server URL**: The endpoint URL of your CrewAI Enterprise server
2. **Bearer Token**: Authentication token for your CrewAI Enterprise API

### Optional Configuration

The Actor automatically handles:
- SSE transport setup
- Error handling and retries
- Charging for operations
- Standby mode configuration

## 📚 Example Usage

### Starting a Crew Task

```python
# Example crew inputs
crew_inputs = {
    "task": "Research and analyze competitor pricing strategies",
    "context": {
        "industry": "SaaS",
        "company_size": "startup",
        "target_market": "SMB"
    },
    "agents": ["researcher", "analyst", "writer"]
}

# Start the crew
result = await kickoff_crew(crew_inputs)
crew_id = result["crew_id"]
```

### Monitoring Task Progress

```python
# Check status periodically
status = await get_crew_status(crew_id)

if status["status"] == "completed":
    print("Task completed!")
    print("Results:", status["results"])
elif status["status"] == "running":
    print("Task still running...")
    print("Progress:", status.get("progress", "Unknown"))
else:
    print("Task status:", status["status"])
```

## 🔗 Related Resources

- [CrewAI Enterprise Documentation](https://docs.crewai.com/enterprise)
- [Model Context Protocol Documentation](https://modelcontextprotocol.io)
- [Apify MCP Documentation](https://docs.apify.com/platform/integrations/mcp)
- [What is Anthropic's Model Context Protocol?](https://blog.apify.com/what-is-model-context-protocol/)

## 🚀 Deployment

### Using Apify CLI

1. Install Apify CLI: `npm install -g apify-cli`
2. Login: `apify login`
3. Deploy: `apify push`

### Using Git Integration

1. Connect your Git repository to Apify
2. Push changes to trigger automatic deployment
3. Configure environment variables in Apify Console

## 🛡️ Security

- Bearer tokens are handled securely through Apify's secret management
- All API communications use HTTPS
- The Actor runs in an isolated container environment
- No sensitive data is logged or stored

## 📞 Support

For issues related to:
- **CrewAI Enterprise**: Contact CrewAI support
- **Apify Platform**: Check [Apify documentation](https://docs.apify.com) or [Discord community](https://discord.com/invite/jyEM2PRvMU)
- **MCP Protocol**: See [MCP documentation](https://modelcontextprotocol.io)

## 📄 License

This Actor is provided as-is under standard Apify terms. CrewAI Enterprise is a separate service with its own licensing terms.