Frequently Asked Questions about UBOS MCP Server

Q: What is an MCP Server?

A: MCP (Model Context Protocol) server is an open protocol that standardizes how applications provide context to LLMs. It acts as a bridge, allowing AI models to access and interact with external data sources and tools.

Q: What are the key features of the UBOS MCP Server?

A: Key features include OpenAI services integration, Git repository analysis, local filesystem operations, Prometheus integration, a unified testing tool, and advanced Git analysis with AI recommendations.

Q: What OpenAI models are supported by the MCP Server?

A: The MCP Server supports both Azure OpenAI and standard OpenAI models, including GPT-4, GPT-3.5 Turbo, and others. You can configure the specific models to use via environment variables.

Q: How do I install the UBOS MCP Server?

A: Clone the repository, install the dependencies using pip install -r requirements.txt, set the required environment variables, and then run the server using python scripts/start_mcp_server.py.

Q: What environment variables do I need to set up the MCP Server?

A: You need to set environment variables for OpenAI (API key, endpoint, deployment name, etc.) and optionally for Prometheus (URL).

Q: How do I test the Git integration?

A: Use the scripts/test_git_integration.py script, providing the URL of the Git repository you want to test.

Q: How can I analyze a Git repository with AI recommendations?

A: Use the scripts/langflow_git_analyzer.py script, providing the URL of the Git repository. You can also use the --search and --diff options for specific analysis.

Q: What Prometheus metrics are useful for monitoring AI applications?

A: Useful PromQL queries include CPU Usage (rate(node_cpu_seconds_total{mode!="idle"}[1m])), Memory Usage (node_memory_MemTotal_bytes - node_memory_MemAvailable_bytes), and Container CPU/Memory Usage (rate(container_cpu_usage_seconds_total[1m]), container_memory_usage_bytes).

Q: How do I use the MCP Server with Langflow?

A: You can use the MCPAIComponent in your Langflow pipelines by providing the MCP server URL. Refer to the documentation for example code snippets.

Q: What are some common troubleshooting steps for Prometheus issues?

A: Verify that Prometheus is running, that you can access the Prometheus UI, that the MCP server is running and accessible, and check the MCP server logs for errors. Start with simple queries to verify connectivity.

Q: How does the MCP Server handle security?

A: The MCP Server provides controlled access to external resources, ensuring that AI models can only interact with data and tools in a secure and authorized manner. Proper configuration and security measures are essential for a secure deployment.

Q: What is the UBOS platform, and how does it relate to the MCP Server?

A: The UBOS platform is a full-stack AI Agent development environment. It amplifies the power of the MCP Server by providing tools for orchestrating AI Agents, connecting to enterprise data, building custom AI Agents, and developing Multi-Agent Systems.