无限画布

Quick start

Requires Node 18+ and a browser. Pick your agent:

Claude Code

claude mcp add rabbithole -- npx -y github:shlokkhemani/rabbithole

Claude Code's default stdio MCP timeout is already long enough for Rabbithole's blocking wait. If you manage .mcp.json manually, this optional per-server field is also fine:

{
  "mcpServers": {
    "rabbithole": {
      "command": "npx",
      "args": ["-y", "github:shlokkhemani/rabbithole"],
      "timeout": 3600000
    }
  }
}

Codex

codex mcp add rabbithole -- npx -y github:shlokkhemani/rabbithole

Then add the timeout field in ~/.codex/config.toml:

[mcp_servers.rabbithole]
command = "npx"
args = ["-y", "github:shlokkhemani/rabbithole"]
tool_timeout_sec = 600

codex mcp add cannot set tool_timeout_sec; edit the config file directly.

Any other MCP client — add this to its MCP config:

{
  "mcpServers": {
    "rabbithole": {
      "command": "npx",
      "args": ["-y", "github:shlokkhemani/rabbithole"]
    }
  }
}

Then start a new agent session and say:

open this document in rabbithole

A browser tab opens. Select any text on the page, ask a question (or tap a lens — Explain · ELI5 · Example · Go Deeper), and the answer streams in as a new document. Keep going.

<details> <summary><strong>Prefer running from a local clone?</strong> (faster startup, easier hacking)</summary>

git clone https://github.com/shlokkhemani/rabbithole.git
cd rabbithole && npm install
claude mcp add rabbithole -- node "$(pwd)/bin/mcp-server.js"

</details>

🤖 If you are an AI agent setting this up

Follow these steps exactly:

1. Check Node: node --version must be ≥ 18. 2. Register the server with the one-liner above that matches the agent you are running inside (claude mcp add … for Claude Code, codex mcp add … for Codex, or write the JSON block into the client's MCP config file). The npx -y github:shlokkhemani/rabbithole form needs no clone and no install step — npx fetches this repo and runs bin/mcp-server.js over stdio. For Codex, also add tool_timeout_sec = 600 under [mcp_servers.rabbithole] in ~/.codex/config.toml; codex mcp add cannot set that field. 3. Verify: after the client reconnects, four tools should be available — open_rabbithole, answer_branch, ingest_pdf, list_rabbitholes. (In Claude Code, claude mcp list should show rabbithole as connected. The server itself also responds to a standard MCP initialize with server name rabbithole.) 4. Tell your human what to do next: they should start a session and say "open &lt;some document&gt; in rabbithole" — you then call open_rabbithole with { title, content } (or { title, file_path }; add base_url when the markdown came from a URL or repo; add assets for local images referenced as asset:name.png), which blocks until they select text and ask. Answer with answer_branch (stream chunks with partial: true, finish with a titled final call) and keep looping until status='session_closed'. If a long wait returns status='keep_listening', immediately call open_rabbithole with the returned { hole_id } and do not re-send content.

Notes for you: the tool call blocks by design (long-poll) — that is normal, not a hang. First npx run takes ~15–30s while it fetches the repo; later runs are cached. If the browser must not auto-open (headless), set RABBITHOLE_NO_BROWSER=1 in the server's env.

Configuration