> ## Documentation Index
> Fetch the complete documentation index at: https://docs.cryptechtoday.com/llms.txt
> Use this file to discover all available pages before exploring further.

# AI & Model Context Protocol (MCP)

> Equip Claude Desktop, Cursor, and autonomous AI agents with native CrypTech token audits

# AI & Model Context Protocol (MCP)

CrypTech provides a native **Model Context Protocol (MCP)** server that allows AI agents running in Claude Desktop, Cursor, or LangChain to audit token safety, verify honeypots, and inspect liquidity directly within AI chat sessions.

***

## What AI Agents Can Do

Once configured, your AI assistant can execute commands like:

* *"Audit the contract `0x514910771AF9Ca656af840dff83E8264EcF986CA` and tell me if it's safe to buy."*
* *"Does this token have an unlocked liquidity pool or high sell tax?"*
* *"Simulate an exit of \$5,000 for this address on Uniswap."*

***

## Claude Desktop Configuration

Open your Claude Desktop configuration file:

* **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json`
* **Windows**: `%APPDATA%\Claude\claude_desktop_config.json`

Add the `cryptech` server configuration:

```json theme={null}
{
  "mcpServers": {
    "cryptech": {
      "command": "node",
      "args": ["/path/to/cryptech/mcp/index.js"],
      "env": {
        "CRYPTECH_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}
```

Restart Claude Desktop. The hammer icon in the prompt input will now display CrypTech tools:

* `analyze_token`
* `get_market_regime`
* `get_alpha_radar`

***

## Cursor IDE Configuration

In your Cursor project workspace, create `.cursor/mcp.json`:

```json theme={null}
{
  "mcpServers": {
    "cryptech": {
      "command": "node",
      "args": ["${workspaceFolder}/mcp/index.js"],
      "env": {
        "CRYPTECH_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}
```

Save the file and refresh Cursor MCP servers in **Settings > Features > MCP**.


## Related topics

- [Introduction](https://docs.cryptechtoday.com/index.md)


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.