DrissionPageMCP
基于DrissionPage和FastMCP的浏览器自动化MCP服务器,提供丰富的浏览器操作API供AI调用
claude mcp add --transport stdio wxhzhwxhzh-drissionpagemcp uv --directory D:\path\to\DrissionPageMCP run main.py
How to use
DrissionPageMCP is a Python-based MCP server that combines DrissionPage with FastMCP to provide browser automation capabilities that can be controlled by AI. It exposes a rich set of browser operations through its MCP API, including opening, closing, and managing browser instances, interacting with page elements, executing JavaScript, using CDP protocol features, handling file downloads, simulating keyboard input, taking screenshots, monitoring network data packets, and auto-uploading/downloading files. This makes it suitable for AI workflows that require automated web interactions and data extraction in a programmable way. To use it, configure the mcpServers entry (see installation notes) and run the server via uv or your preferred MCP launcher. The server will expose APIs that your client can call to perform the browser automation tasks described above.
How to install
Prerequisites:
- Python 3.9 or newer
- pip (latest)
- uv (latest, used to run MCP servers via Python scripts)
- Chrome browser (recommended 90+)
- Internet access to install Python packages
Installation steps:
- Clone or download the repository to your local machine.
- Create and activate a Python virtual environment (recommended):
- python -m venv venv
- On Windows: venv\Scripts\activate
- On macOS/Linux: source venv/bin/activate
- Install required Python packages (adjust versions as needed):
- pip install --upgrade pip
- pip install drissionpage>=4.1.0.18 fastmcp>=2.4.0 uv
- Ensure the main application file (main.py) is present in the project directory (as indicated in the README).
- Verify that the uv command is available in your PATH (the MCP runner will use it to start the server).
- Prepare your mcpServers configuration (see mcp_config example) and place it in your editor or MCP manager.
Running locally:
- Use the provided run command (example shown in mcp_config): uv run path\to\DrissionPageMCP\main.py
- Alternatively, from the editor/IDE, ensure the mcpServers entry points to the same main.py path.
Additional notes
Tips and common issues:
- Always set the correct absolute path to your main.py file in the mcpServers configuration and escape Windows paths (\).
- Ensure uv is installed and accessible in your system PATH before attempting to run the server.
- If you modify dependencies or Python version, re-create the virtual environment and reinstall packages.
- The MCP configuration provided in the README uses the type stdio; adjust the type only if your editor requires a different integration approach.
- For debugging, you can use commands like: npx -y @modelcontextprotocol/inspector uv run D:\path\to\DrissionPageMCP\main.py or mcp dev D:\path\to\DrissionPageMCP\main.py
- No npm package is required for this Python-based MCP server (npm_package is null).
Related MCP Servers
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
mcp-yfinance
Real-time stock API with Python, MCP server example, yfinance stock analysis dashboard
cloudwatch-logs
MCP server from serkanh/cloudwatch-logs-mcp
DrissionPage
The Model Context Protocol (MCP) service based on DrissionPage provides AI assistants with powerful browser automation capabilities. | 基于 DrissionPage 的模型上下文协议 (MCP) 服务,为 AI 助手提供强大的浏览器自动化能力。
servicenow-api
ServiceNow MCP Server and API Wrapper
the -company
TheMCPCompany: Creating General-purpose Agents with Task-specific Tools