MCP Tool Poisoning Attacks MCP Server

1. install VSCode, Docker 2. install VSCode Extensions *Dev ContainerS* 3. On the VSCode, `Ctrl shift p` and run `Dev Containers: Rebuild Containers`

Local serverstdioTypeScript

What is the MCP Tool Poisoning Attacks MCP server?

Mcp tool poisoning attacks mcp server lets Claude, Cursor and other MCP clients work with MCP Tool Poisoning Attacks directly. 1. install VSCode, Docker 2. install VSCode Extensions Dev ContainerS 3. On the VSCode, Ctrl shift p and run Dev Containers: Rebuild Containers 4. create .env and add ANTHROPIC_API_KEY.

What MCP Tool Poisoning Attacks does

  1. install VSCode, Docker 2. install VSCode Extensions Dev ContainerS 3. On the VSCode, Ctrl shift p and run Dev Containers: Rebuild Containers 4. create .env and add ANTHROPIC_API_KEY

Tools it exposes

Once connected, the assistant can call this tool directly:

  • claude-3-5-haiku — 20241022

Installing the mcp tool poisoning attacks mcp server

Setup follows the standard MCP pattern: clone or install the server, then register it in your client's configuration file and restart the client. The configuration snippets on this page cover Claude Desktop, Claude Code and Cursor.

Configuration

Before the server will start you need to supply one environment variable: ANTHROPIC_API_KEY. Keep credentials in your client's env block or a secrets manager rather than committing them.

Where it fits

Cloud and DevOps servers put infrastructure on speaking terms with your assistant, so "why is this broken?" is answered from real state rather than a generic checklist. MCP Tool Poisoning Attacks sits in that group, and the shape of its toolset — claude-3-5-haiku — tells you what it is really for. Worth comparing against the other cloud devops servers in this directory before you commit to one, since several overlap in scope but differ sharply in setup cost and permissions.

Practical notes

  • This server runs locally, so it operates with whatever access your machine and its credentials already have. Scope that deliberately rather than by default.
  • It will not start until its required credentials are present, so set those before wondering why the tools never appear.
  • Maintained by RyosukeDTomita, written in TypeScript.
  • MCP clients ask for confirmation before each tool call by default. Keep that on while you learn what the mcp tool poisoning attacks mcp server actually does with your data.
  • Every entry in this directory is reviewed by hand before it goes live, and details are checked against the project's own documentation.

Available tools

ToolWhat it does
claude-3-5-haiku20241022

How to install the MCP Tool Poisoning Attacks MCP server

{
  "mcpServers": {
    "ipinfo": {
      "command": "/usr/local/bin/npx",
      "args": [
        "-y",
        "github:RyosukeDTomita/mcp_tool_poisoning_attacks#main",
        "ipinfo"
      ],
      "env": {
        "PATH": "/usr/local/bin:/usr/bin:/bin"
      }
    },
    "burp": {
      "url": "http://localhost:9876/sse"
    }
  }
}

Configuration as documented by the project. Restart the client after saving.

Configuration

VariableDescriptionRequired
ANTHROPIC_API_KEYCredential the server authenticates with.Yes

Example prompts to try

  • Use MCP Tool Poisoning Attacks to claude-3-5-haiku.

Frequently asked questions

It connects MCP Tool Poisoning Attacks to MCP-compatible AI assistants such as Claude and Cursor, exposing 1 tool (claude-3-5-haiku) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with MCP Tool Poisoning Attacks directly.