For AI apps

Use Toll in your AI

Add Toll to the AI app you use, then ask it about a Robinhood Chain coin or wallet. It reads the chain through Toll and links you to the full page. Free, no account, no key.

On this page

The server address is the same everywhere:

https://t0ll.xyz/mcp

It is a remote MCP server (Model Context Protocol, Streamable HTTP). It has no sign-in.

What it does

Tool	You give it	You get
check_coin	a token address	Where and when it launched, the maker and its launch record, snipers, bundles (wallets exempt from tax), flagged wallets, what the maker holds, the curve or pools, and our findings.
dev_record	a wallet	Every coin the wallet launched on Pons and Long, how many graduated, and whether it sold.
wallet_privacy	a wallet	A privacy grade, a score and each leak: what anyone can learn about the wallet from the chain.
fresh_launches	how many (up to 20)	The newest coins on the chain, with our live engine's notes on each.
chain_pulse	nothing	Launches, graduations, new pools and trades right now.

Try asking: “Check 0x… on Robinhood Chain”, “Who made this coin and what else did they launch?”, “What can people see about my wallet?” or “What launched in the last few minutes?”

Claude Code

Run this in a terminal:

claude mcp add --transport http toll https://t0ll.xyz/mcp

Add --scope user to have it in every project.

Claude desktop and claude.ai

Open Settings, then Connectors. In newer versions it is under Customize.
Choose Add custom connector.
Name it Toll, paste https://t0ll.xyz/mcp and add it.
In a chat, turn Toll on from the tools menu.

The free plan allows one custom connector.

ChatGPT

On the web, open Settings, then Apps, then Advanced settings, and turn on Developer mode. In some versions it is under Security and login.
Create a new app (connector) with the URL https://t0ll.xyz/mcp and authentication No authentication.
In a chat, pick it from the developer mode tools.

Developer mode is on Plus, Pro, Business, Enterprise and Edu, not the free plan. To build a custom GPT instead, import /openapi.json as its action.

Cursor

Add this to ~/.cursor/mcp.json (every project) or .cursor/mcp.json (one project):

{
  "mcpServers": {
    "toll": {
      "url": "https://t0ll.xyz/mcp"
    }
  }
}

VS Code (Copilot)

Add this to .vscode/mcp.json in your project, then start the server from that file or with MCP: List Servers:

{
  "servers": {
    "toll": {
      "type": "http",
      "url": "https://t0ll.xyz/mcp"
    }
  }
}

Gemini CLI

Run:

gemini mcp add --transport http toll https://t0ll.xyz/mcp

Or add this to ~/.gemini/settings.json:

{
  "mcpServers": {
    "toll": {
      "httpUrl": "https://t0ll.xyz/mcp"
    }
  }
}

Use httpUrl, not url, which Gemini CLI reads as the older SSE transport.

Windsurf

Add this to its mcp_config.json (Settings, then Cascade, then MCP servers, then View raw config):

{
  "mcpServers": {
    "toll": {
      "serverUrl": "https://t0ll.xyz/mcp"
    }
  }
}

Other apps

Any MCP client that takes a remote server: https://t0ll.xyz/mcp, transport Streamable HTTP, no authentication.
Server card: /.well-known/mcp.json.
OpenAPI 3.1 for GPT Actions and LangChain: /openapi.json (the free check, deployer and privacy reads).
Agents that pay per call: Toll for agents (x402).

Notes

Robinhood Chain only. Other chains are not read.
Every answer is facts read from the chain. It ends with a link to the full page on t0ll.xyz.
A coin or wallet nobody has asked about takes longer to read the first time. The answer then says to try again in a few seconds.
20 tool calls a minute per connection. Fresh chain reads share the site’s budget, so at busy times the answer may ask you to wait a minute.
Information only. Nothing here is financial advice, and nothing here trades for you.