Frequently Asked Questions about GroundDocs

Q: What is GroundDocs?

A: GroundDocs is a version-aware documentation assistant for Large Language Models (LLMs). It connects LLMs to trusted, real-time documentation, reducing hallucinations and ensuring accurate, version-specific responses.

Q: How does GroundDocs reduce LLM hallucinations?

A: GroundDocs connects LLMs to trusted documentation sources, allowing them to verify information and provide responses based on factual data. This reduces the reliance on the LLM’s pre-trained knowledge, which may be outdated or inaccurate.

Q: What is an MCP Server and how does GroundDocs utilize it?

A: MCP stands for Model Context Protocol. An MCP server acts as a bridge, allowing AI models to access and interact with external data sources and tools. GroundDocs utilizes a local MCP server to handle inference-time queries and retrieve information from a remote backend data repository.

Q: What kind of documentation does GroundDocs support?

A: Currently, GroundDocs supports Kubernetes documentation (all versions, including version-aware kubectl behavior, API schemas, and feature gates). Support for additional domains is planned for the future.

Q: How do I install GroundDocs?

A: You can install GroundDocs using the following command:

bash npx @grounddocs/cli@latest install

Replace <client> with your IDE or tool. Supported clients include cursor, windsurf, cline, claude, witsy, enconvo, and vscode.

Q: Can I configure GroundDocs manually?

A: Yes, you can manually configure GroundDocs by adding it to your IDE’s MCP configuration. Refer to the documentation for detailed instructions.

Q: What is the architecture of GroundDocs?

A: GroundDocs consists of a local MCP server (lightweight, public, runs inference-time queries) and a remote backend data repository (private, handles scraping, indexing, and heavy lifting).

Q: How does GroundDocs integrate with UBOS?

A: GroundDocs is seamlessly integrated into the UBOS full-stack AI Agent development platform. It allows UBOS users to build more accurate and reliable AI Agents by providing them with access to the latest documentation.

Q: Can I contribute to GroundDocs?

A: Yes, contributions are welcome! Please feel free to submit a Pull Request.

Q: What are the benefits of using GroundDocs with UBOS?

A: Using GroundDocs with UBOS allows you to:

• Reduce LLM hallucinations.
• Improve the accuracy of AI Agents.
• Access version-specific documentation.
• Build more trustworthy AI applications.

Q: What if my IDE is not listed as a supported client?

A: Even if your IDE isn’t explicitly listed, you can often use the manual configuration option by adding GroundDocs to your IDE’s MCP configuration. Refer to your IDE’s documentation for instructions on configuring MCP servers.