Skip to main content
Agents operating on CNAP need context — how products work, what API endpoints are available, how to configure deployments. The documentation MCP server gives them real-time access to the full CNAP docs, with no authentication required. 
https://docs.cnap.tech/mcp
This is the documentation MCP server for searching docs. For the platform MCP server that manages CNAP resources, see Platform MCP.

​
What It Does

  • Real-time search — AI tools query CNAP documentation as you chat, returning up-to-date answers
  • Zero configuration — no API keys, no OAuth, no auth at all
  • Any MCP client — works with Claude, Cursor, VS Code, and any tool that supports MCP

​
Try It

Once connected (setup instructions), ask your AI assistant:
Example prompts
How do I create a new product in CNAP?
What's the difference between managed and imported clusters?
How do I set up pricing for my products?
The MCP server returns real-time data from the latest documentation — no stale training data.

​
Other Ways AI Tools Access Docs

Beyond the MCP server, CNAP provides multiple channels for AI tools:

​
Contextual Menu

Every documentation page includes a contextual menu with quick actions:
  • Copy page — copy content as Markdown for pasting into AI tools
  • View as Markdown — open the page in raw Markdown format
  • Open in AI tools — start a conversation in ChatGPT, Claude, or Perplexity with page context
  • Connect to editors — one-click MCP setup for Cursor or VS Code

​
LLMs.txt Files

Industry-standard files for AI indexing:
  • llms.txt — all pages with descriptions for quick reference
  • llms-full.txt — complete documentation in one file

​
Markdown Access

Append .md to any URL for AI-friendly Markdown output: 
https://docs.cnap.tech/ai/docs-mcp.md

​
Troubleshooting

  • Verify URL: https://docs.cnap.tech/mcp
  • Check internet connection
  • Ensure your AI tool supports the MCP protocol
  • Try the contextual menu on any doc page for automatic setup
  • Check JSON syntax in config files
  • Verify file permissions and correct file location
  • Restart your AI application after config changes
  • The MCP server always returns real-time data
  • Restart the MCP connection in your tool
  • Clear cached responses in your AI app

Quick Setup

Configure your AI tool to connect in under a minute

Platform MCP

Manage infrastructure and deployments through AI agents