Skip to main content
Every page in the Kameleoon documentation includes built-in options for connecting content to your AI tools. Use them to ask questions about an individual article in one click, or configure a persistent MCP server that gives your IDE or AI agent access to the full documentation.

​
Per-page options

A contextual menu at the top of each article provides the following actions:

​
Configure a persistent docs MCP server

For ongoing use, connect the Kameleoon docs as a persistent MCP server so your AI assistant can answer documentation questions without copying individual pages each time.

​
Step 1: Copy the MCP server configuration

Open any page in the Kameleoon docs and click Copy MCP Server from the contextual menu at the top of the article. Clicking this option copies a pre-configured snippet for the Kameleoon docs MCP server.

​
Step 2: Add the server to your tool

  1. Open your Claude Desktop configuration file:
    • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
    • Windows: %APPDATA%\Claude\claude_desktop_config.json
  2. Paste the copied snippet into the mcpServers object.
  3. Save the file and restart Claude Desktop.
After restarting, start a new conversation and ask a documentation question, for example: “How do I set up a feature flag in the Python SDK?” or “What statistical methods does Kameleoon support?”

​
What the docs MCP server can do

After you connect the Kameleoon docs MCP server, your AI assistant can:
  • Answer questions about any topic covered in the Kameleoon documentation.
  • Retrieve and summarize articles from the user manual and developer docs.
  • Walk you through setup guides, SDK references, and API documentation step by step.
The docs MCP server provides read access to documentation. To manage feature flags, retrieve experiment results, or automate your implementation workflow from an AI assistant, see the Kameleoon MCP server.