Manus MCP
by huyouare
Manus MCP is a Model Context Protocol (MCP) server implementation. It enables browsing the web, performing search queries, and executing code.
Last updated: N/A
What is Manus MCP?
Manus MCP is a server that implements the Model Context Protocol, allowing it to interact with tools like web browsers, search engines, and code interpreters. It provides a set of tools that can be used by AI models to perform various tasks.
How to use Manus MCP?
To use Manus MCP, you need to configure it within a compatible application like Claude for Desktop by specifying the command and arguments required to run the server. The README provides a sample configuration for Claude for Desktop. After configuration, the tools provided by Manus MCP will be available within the application.
Key features of Manus MCP
Google Search Tool
Web Browsing Tool (navigate, click, input_text, get_content, execute_js, scroll, refresh)
Code Interpreter Tool (read, write, execute, list) supporting multiple languages
Bash Tool for executing shell commands
Use cases of Manus MCP
Web research and information gathering
Automated web interaction and data extraction
Code execution and testing in a sandboxed environment
Performing system administration tasks via shell commands
Integrating external tools with AI models
FAQ from Manus MCP
What is the default sandbox directory?
What is the default sandbox directory?
The default sandbox directory is ~/manus-sandbox
.
What is the default global timeout?
What is the default global timeout?
The default global timeout is 60 seconds.
How do I run the browser in headless mode?
How do I run the browser in headless mode?
Set the BROWSER_HEADLESS
environment variable to true
.
How do I change the maximum number of Google search results?
How do I change the maximum number of Google search results?
Set the GOOGLE_SEARCH_MAX_RESULTS
environment variable to the desired number.
What programming languages are supported by the code interpreter?
What programming languages are supported by the code interpreter?
The code interpreter supports multiple programming languages including Python, JavaScript (Node.js), Bash, Ruby, Perl, and R.