Frequently Asked Questions about MCP Servers and UBOS

Q: What exactly is an MCP Server?

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

Q: How does Clarity MCP work?

A: Clarity MCP uses HAR (HTTP Archive) files to capture browser network requests, extracts API endpoints, and then uses AI to generate tools that AI agents can use to interact with those APIs.

Q: What are the benefits of using MCP Servers?

A: MCP Servers enable AI agents to access real-time information, automate tasks, personalize recommendations, and integrate with various digital services, improving efficiency and accuracy.

Q: What is a HAR file and how do I create one?

A: A HAR (HTTP Archive) file captures all network requests your browser makes while interacting with a website. You can create one using your browser’s developer tools (usually accessed by pressing F12).

Q: How do I install and run Clarity MCP?

A: You can install Clarity MCP using npm (npm install -g @lekt9/clarity-mcp). After setting your API key, you can run the server with the clarity-mcp command.

Q: What is UBOS, and how does it relate to MCP Servers?

A: UBOS is a full-stack AI Agent Development Platform that helps you orchestrate AI Agents, connect them with your enterprise data, and build custom AI Agents. MCP Servers can be integrated into the UBOS platform to enhance AI agent capabilities.

Q: What kind of information can be found in HAR files and what are the security considerations?

A: HAR files can contain sensitive information like cookies, authentication tokens, and personal data. Handle them securely and avoid sharing them publicly.

Q: Can I use Clarity MCP with AI models other than Claude?

A: While the documentation specifically mentions Claude, the underlying principles of MCPs and the OpenAPI compliance of Clarity MCP suggest it can be adapted for use with other AI models that support the MCP protocol.

Q: Are MCPs just APIs with a different name?

A: MCPs can be considered APIs packaged in a way that’s optimized for AI consumption. They provide a standardized way for AI agents to interact with external services.

Q: What are the limitations of Clarity MCP?

A: The server requires a valid API key, and tool execution depends on the availability of the Clarity backend. It also does not cache tool definitions or results.