Frequently Asked Questions (FAQ)

Q: What is the Model Context Protocol (MCP)? A: MCP is an open protocol that standardizes how applications provide context to Large Language Models (LLMs), enabling seamless interaction between LLMs and external data sources and tools.

Q: What is the UBOS MCP Server? A: The UBOS MCP Server is an implementation of the MCP protocol that provides tools for LLMs to interact with web content, such as web scraping.

Q: How do I install the UBOS MCP Server? A: You can install the UBOS MCP Server by cloning the repository, navigating to the directory, and running pip install -r requirements.txt.

Q: How do I run the UBOS MCP Server? A: You can run the server using the provided run.sh or run.bat scripts, or by running python server.py directly.

Q: How do I integrate the UBOS MCP Server with Claude for Desktop? A: You can integrate the server with Claude by adding a configuration block to the claude_desktop_config.json file.

Q: What is the web_scrape tool? A: The web_scrape tool fetches the content of a specified URL and returns it in markdown format.

Q: How can I add more tools to the UBOS MCP Server? A: You can add more tools by creating new Python files in the tools/ directory and registering them with the MCP server.

Q: What is the Streamlit UI used for? A: The Streamlit UI allows you to connect to and test all your MCP servers configured in Claude for Desktop.

Q: What do I do if I have missing dependencies? A: Make sure all dependencies in requirements.txt are installed using pip install -r requirements.txt.

Q: What do I do if I have connection issues? A: Check that the server is running and that the configuration in Claude for Desktop points to the correct path.

Q: Is the UBOS MCP Server part of a larger platform? A: Yes, the UBOS MCP Server is a component of the UBOS platform, a full-stack AI Agent Development Platform.