Skip to content

Use with AI

Connect your AI tool to our Model Context Protocol (MCP) server so it can search and read the developer documentation while helping you build an integration.

The server is available at /mcp on this documentation site:

https://docs.conjuringlab.com/mcp

Use Streamable HTTP as the transport. No account, API key, or authentication is required. The server provides read-only access to the documentation.

Add the server with the Codex CLI:

Terminal window
codex mcp add conjuringlab-docs --url https://docs.conjuringlab.com/mcp

Run codex mcp list to confirm it is configured. See the Codex MCP setup guide for more options.

Run this command to make the server available across your projects:

Terminal window
claude mcp add --transport http --scope user conjuringlab-docs https://docs.conjuringlab.com/mcp

In Claude Code, use /mcp to check the connection. See Claude Code’s MCP documentation for more options.

Add the following server to .cursor/mcp.json in your project. If the file already has an mcpServers object, add conjuringlab-docs alongside the existing servers.

.cursor/mcp.json
{
"mcpServers": {
"conjuringlab-docs": {
"url": "https://docs.conjuringlab.com/mcp"
}
}
}

To use it across all projects, put the same configuration in ~/.cursor/mcp.json. See Cursor’s MCP documentation for more options.

In your tool’s MCP settings, add a remote server named conjuringlab-docs, paste the endpoint URL above, choose Streamable HTTP, and leave authentication disabled.

Once connected, ask your AI tool to use the Conjuring Lab documentation. For example:

Use the Conjuring Lab docs to write a Swift function that sends text to Motherboard X through its URL scheme.

Check the Conjuring Lab docs and show me how to change an Akronym user’s default word, including authentication and error handling.

The server exposes three tools:

Tool Purpose
search_docs Find documentation matching a query.
get_doc Read a documentation page by its path.
list_docs List the available documentation pages.