mcpField
A
AI Meta MCP Server

AI Meta MCP Server

Local
github.com/alxspikerAI & Agents10Updated 65w ago
Repository

About AI Meta MCP Server

AI Meta MCP Server lets AI models define and run custom tools at runtime through MCP. It is intended for extending tool capabilities on demand, with sandboxed execution, human approval, and persistent tool definitions.

Key features

  1. 01Create custom tools dynamically
  2. 02Run JavaScript, Python, and Shell
  3. 03Execute in sandboxes
  4. 04Persist tool definitions
  5. 05Manage the tool registry
  6. 06Require approval to create and run

Use cases

  1. 01Create calculation tools on demand
  2. 02Reuse custom tools across sessions
  3. 03Run generated code after approval

MCP support

The MCP server exposes meta-tools for defining and executing custom functions, and lets MCP clients manage those tools. Code runs in sandboxes, with human approval required for creation and execution.

#ai-meta-mcp-server#dynamic-tools#meta-function-architecture

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "ai-meta": {
      "command": "npx",
      "args": [
        "-y",
        "ai-meta-mcp-server"
      ]
    }
  }
}
Run commandnpx -y ai-meta-mcp-server

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
npmai-meta-mcp-server

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

From the project README

AI Meta MCP Server

A dynamic MCP server that allows AI models to create and execute their own custom tools through a meta-function architecture. This server provides a mechanism for AI to extend its own capabilities by defining custom functions at runtime.

Features

  • Dynamic Tool Creation: AI can define new tools with custom implementations
  • Multiple Runtime Environments: Support for JavaScript, Python, and Shell execution
  • Sandboxed Security: Tools run in isolated sandboxes for safety
  • Persistence: Store and load custom tool definitions between sessions
  • Flexible Tool Registry: Manage, list, update, and delete custom tools
  • Human Approval Flow: Requires explicit human approval for tool creation and execution

Security Considerations

⚠️ WARNING: This server allows for dynamic code execution. Use with caution and only in trusted environments.

  • All code executes in sandboxed environments
  • Human-in-the-loop approval required for tool creation and execution
  • Tool execution privileges configurable through environment variables
  • Audit logging for all operations

Installation

npm install ai-meta-mcp-server

Usage

Running the server
npx ai-meta-mcp-server
Running with Docker
# Build the Docker image
docker build -t ai-meta-mcp-server .

# Run the container
docker run --rm -i ai-meta-mcp-server

# Run with custom configuration and persistent storage
docker run --rm -i \
  -e ALLOW_PYTHON_EXECUTION=true \
  -e ALLOW_SHELL_EXECUTION=false \
  -v $(pwd)/data:/app/data \
  ai-meta-mcp-server
Configuration

Environment variables:

  • ALLOW_JS_EXECUTION: Enable JavaScript execution (default: true)
  • ALLOW_PYTHON_EXECUTION: Enable Python execution (default: false)
  • ALLOW_SHELL_EXECUTION: Enable Shell execution (default: false)
  • PERSIST_TOOLS: Save tools between sessions (default: true)
  • TOOLS_DB_PATH: Path to store tools database (default: "./tools.json")
Running with Claude Desktop

Add this to your claude_desktop_config.json:

{
  "mcpServers": {
    "ai-meta-mcp": {
      "command": "npx",
      "args": ["-y", "ai-meta-mcp-server"],
      "env": {
        "ALLOW_JS_EXECUTION": "true",
        "ALLOW_PYTHON_EXECUTION": "false",
        "ALLOW_SHELL_EXECUTION": "false"
      }
    }
  }
}

Tool Creation Example

In Claude Desktop, you can create a new tool like this:

Can you create a tool called "calculate_compound_interest" that computes compound interest given principal, rate, time, and compounding frequency?

Claude will use the define_function meta-tool to create your new tool, which becomes available for immediate use.

Architecture

The server implements the Model Context Protocol (MCP) and provides a meta-tool architecture that enables AI-driven function registration and execution within safe boundaries.

License

MIT

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.