A Model Context Protocol (MCP) server implementation that can browse the web, perform search queries, and execute code.
To use Manus MCP with Claude for Desktop:
Create or edit the Claude for Desktop configuration file:
~/Library/Application Support/Claude/claude_desktop_config.json
Add the following configuration:
{
"mcpServers": {
"manus-mcp": {
"command": "uv",
"args": [
"--directory",
"/ABSOLUTE/PATH/TO/manus-mcp",
"run",
"mcp_server.py"
]
}
}
}
Restart Claude for Desktop
You should now see the Manus MCP tools available in Claude for Desktop
A simple greeting tool that returns a welcome message.
Performs Google searches and returns a list of relevant links.
Interacts with a web browser to navigate websites and extract information. Supported actions:
navigate
: Go to a specific URLclick
: Click an element by indexinput_text
: Input text into an elementget_content
: Get the page contentexecute_js
: Execute JavaScript codescroll
: Scroll the pagerefresh
: Refresh the current pageAllows reading, writing, and executing code files in a sandboxed environment. Supported actions:
read
: Read the contents of a filewrite
: Write content to a fileexecute
: Execute a file or code snippetlist
: List files in the sandboxSupports multiple programming languages including Python, JavaScript (Node.js), Bash, Ruby, Perl, and R.
Executes bash commands in the sandbox directory. Features:
The following environment variables can be configured:
SANDBOX_DIR
: Path to the sandbox directory (default: ~/manus-sandbox
)GLOBAL_TIMEOUT
: Global timeout for all operations in seconds (default: 60)BROWSER_HEADLESS
: Whether to run the browser in headless mode (default: false)GOOGLE_SEARCH_MAX_RESULTS
: Maximum number of search results to return (default: 10)LOG_LEVEL
: Logging level (default: INFO)./setup.sh
Clone the repository
git clone https://github.com/yourusername/manus-mcp.git
cd manus-mcp
Create a virtual environment and install dependencies
uv venv
source .venv/bin/activate
uv pip install -e . # Install the project and its dependencies
Run the server
# Make sure your virtual environment is activated
source .venv/bin/activate
./run.py
# or
uvicorn app.main:app --reload
Visit http://localhost:8000/docs
to see the API documentation
To install development dependencies:
uv pip install -e ".[dev]"
The API follows the Model Context Protocol (MCP) specification.
{
"mcpServers": {
"manus-mcp": {
"env": {},
"args": [
"--directory",
"/ABSOLUTE/PATH/TO/manus-mcp",
"run",
"mcp_server.py"
],
"command": "uv"
}
}
}
Seamless access to top MCP servers powering the future of AI integration.