Read these docs with your agent

Connect Claude, Cursor, Codex, Gemini, or any MCP client to the DataGenie docs and let your agent read the same structured content you see here — in under 10 seconds.

These docs are agent-native. The same structured content you read here is served to AI agents over a first-class MCP server — author once, humans read the pages, agents query the corpus. No scraping, no copy-paste, no drift.

One endpoint, public, no key — connect in under 10 seconds

Point any MCP client at https://docs.datagenie.ai/mcp. It's a public, unauthenticated Streamable HTTP endpoint — paste the URL, and your agent can search and read every page, structured.

Connect your client

Add the server

Use the snippet for your client below. The only value you ever need is the endpoint URL: https://docs.datagenie.ai/mcp.

Reload

Restart the client (or reload its MCP config). The datagenie-docs server appears with five tools.

Ask

Ask your agent something like "Using the datagenie-docs MCP, how does the Nirvana algorithm join data without ETL?" — it reads the answer straight from these docs.

Add the server from your terminal:

claude mcp add --transport http datagenie-docs https://docs.datagenie.ai/mcp

Add --scope user to make it available in every project on your machine:

claude mcp add --transport http --scope user datagenie-docs https://docs.datagenie.ai/mcp

Verify with claude mcp list — datagenie-docs should report Connected.

In the Claude app or on claude.ai:

Open Settings → Connectors.
Click Add custom connector.
Name it DataGenie Docs and set the URL to https://docs.datagenie.ai/mcp.
Save — Claude can now read the docs in any conversation.

Custom connectors are available on Claude Pro, Max, Team, and Enterprise plans.

Add an entry to ~/.codex/config.toml:

[mcp_servers.datagenie_docs]
command = "npx"
args = ["-y", "mcp-remote", "https://docs.datagenie.ai/mcp"]

The mcp-remote bridge connects Codex's stdio transport to the remote HTTP server and works on every Codex version. Restart Codex to pick up the change.

Add the server to ~/.gemini/settings.json:

{
  "mcpServers": {
    "datagenie-docs": {
      "httpUrl": "https://docs.datagenie.ai/mcp"
    }
  }
}

Use httpUrl for the Streamable HTTP transport. Run /mcp inside Gemini CLI to confirm the server is listed.

Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (per project):

{
  "mcpServers": {
    "datagenie-docs": {
      "url": "https://docs.datagenie.ai/mcp"
    }
  }
}

Cursor picks up the server automatically — check Settings → MCP for a green status dot.

For GitHub Copilot agent mode, add .vscode/mcp.json to your workspace:

{
  "servers": {
    "datagenie-docs": {
      "type": "http",
      "url": "https://docs.datagenie.ai/mcp"
    }
  }
}

VS Code uses servers (not mcpServers) with an explicit type.

The endpoint speaks standard MCP over Streamable HTTP with no authentication. Most clients accept this shape:

{
  "mcpServers": {
    "datagenie-docs": {
      "type": "http",
      "url": "https://docs.datagenie.ai/mcp"
    }
  }
}

If your client only supports stdio, bridge it with npx -y mcp-remote https://docs.datagenie.ai/mcp.

What your agent can do

Once connected, the agent has five tools over the entire docs corpus — the same structured content the site renders, not flattened HTML.

Tool	What it does
`search`	Full-text search across the knowledge base; returns ranked page hits.
`get_page`	A page's full structured content by path.
`list`	List knowledge nodes, filtered by kind, tag, or entity type.
`navigate`	Follow relations from a page path, node id, or entity reference.
`get_component_data`	The structured data behind a rendered component — its knowledge face.

Every component on these pages — the architecture canvases, the changelog timeline, the comparison tables — serializes to typed data the agent reads directly. Author once; humans and agents get the same corpus.

Read these docs with your agent

Connect your client

Add the server

Reload

Ask

What your agent can do

On this page