Crawlab MCP Server

Created By

crawlab-team9 months ago

Content

Crawlab MCP Server

This is a Model Context Protocol (MCP) server for Crawlab, allowing AI applications to interact with Crawlab's functionality.

Overview

The MCP server provides a standardized way for AI applications to access Crawlab's features, including:

Spider management (create, read, update, delete)
Task management (run, cancel, restart)
File management (read, write)
Resource access (spiders, tasks)

Architecture

The MCP Server/Client architecture facilitates communication between AI applications and Crawlab:

graph TB
    User[User] --> Client[MCP Client]
    Client --> LLM[LLM Provider]
    Client <--> Server[MCP Server]
    Server <--> Crawlab[Crawlab API]

    subgraph "MCP System"
        Client
        Server
    end

    subgraph "Crawlab System"
        Crawlab
        DB[(Database)]
        Crawlab <--> DB
    end

    class User,LLM,Crawlab,DB external;
    class Client,Server internal;

    %% Flow annotations
    LLM -.-> |Tool calls| Client
    Client -.-> |Executes tool calls| Server
    Server -.-> |API requests| Crawlab
    Crawlab -.-> |API responses| Server
    Server -.-> |Tool results| Client
    Client -.-> |Human-readable response| User

    classDef external fill:#f9f9f9,stroke:#333,stroke-width:1px;
    classDef internal fill:#d9edf7,stroke:#31708f,stroke-width:1px;

Communication Flow

User Query: The user sends a natural language query to the MCP Client
LLM Processing: The Client forwards the query to an LLM provider (e.g., Claude, OpenAI)
Tool Selection: The LLM identifies necessary tools and generates tool calls
Tool Execution: The Client sends tool calls to the MCP Server
API Interaction: The Server executes the corresponding Crawlab API requests
Response Generation: Results flow back through the Server to the Client to the LLM
User Response: The Client delivers the final human-readable response to the user

Installation and Usage

Option 1: Install as a Python package

You can install the MCP server as a Python package, which provides a convenient CLI:

# Install from source
pip install -e .

# Or install from GitHub (when available)
# pip install git+https://github.com/crawlab-team/crawlab-mcp-server.git

After installation, you can use the CLI:

# Start the MCP server
crawlab_mcp-mcp server [--spec PATH_TO_SPEC] [--host HOST] [--port PORT]

# Start the MCP client
crawlab_mcp-mcp client SERVER_URL

Option 2: Running Locally

Prerequisites

Python 3.8+
Crawlab instance running and accessible
API token from Crawlab

Configuration

Copy the .env.example file to .env:
```
cp .env.example .env
```

Edit the .env file with your Crawlab API details:

CRAWLAB_API_BASE_URL=http://your-crawlab-instance:8080/api
CRAWLAB_API_TOKEN=your_api_token_here

Running Locally

Install dependencies:
```
pip install -r requirements.txt
```
Run the server:
```
python server.py
```

Running with Docker

Build the Docker image:
```
docker build -t crawlab-mcp-server .
```

Run the container:

docker run -p 8000:8000 --env-file .env crawlab-mcp-server

Integration with Docker Compose

To add the MCP server to your existing Crawlab Docker Compose setup, add the following service to your docker-compose.yml:

services:
  # ... existing Crawlab services
  
  mcp-server:
    build: ./backend/mcp-server
    ports:
      - "8000:8000"
    environment:
      - CRAWLAB_API_BASE_URL=http://backend:8000/api
      - CRAWLAB_API_TOKEN=your_api_token_here
    depends_on:
      - backend

Using with AI Applications

The MCP server enables AI applications to interact with Crawlab through natural language. Following the architecture diagram above, here's how to use the MCP system:

Setting Up the Connection

Start the MCP Server: Make sure your MCP server is running and accessible
Configure the AI Client: Connect your AI application to the MCP server

Example: Using with Claude Desktop

Open Claude Desktop
Go to Settings > MCP Servers
Add a new server with the URL of your MCP server (e.g., http://localhost:8000)
In a conversation with Claude, you can now use Crawlab functionality by describing what you want to do in natural language

Example Interactions

Based on our architecture, here are example interactions with the system:

Create a Spider:

User: "Create a new spider named 'Product Scraper' for the e-commerce project"
↓
LLM identifies intent and calls the create_spider tool
↓
MCP Server executes the API call to Crawlab
↓
Spider is created and details are returned to the user

Run a Task:

User: "Run the 'Product Scraper' spider on all available nodes"
↓
LLM calls the run_spider tool with appropriate parameters
↓
MCP Server sends the command to Crawlab API
↓
Task is started and confirmation is returned to the user