A

AI Meta MCP Server

Enables AI models to dynamically create and execute their own custom tools through a meta-function a...

61 views
0 installs
Updated Feb 5, 2026
Not audited
Tools I Recommend
Enables AI models to dynamically create and execute their own custom tools through a meta-function architecture, supporting JavaScript, Python, and Shell runtimes with sandboxed security and human approval flows.
  1. Click the "Install Now" button to open the FastMCP connection interface where you add environment variables for this MCP server.
  2. In the FastMCP connection modal, find the "Environment Variables" (or "ENV") section where you can add key/value pairs.
  3. Add the following variables (use the string values shown). These names and defaults come from the README—enter the exact key names below:
    1. ALLOW_JS_EXECUTION = "true"
      • Enables JavaScript runtime (recommended default).
    2. ALLOW_PYTHON_EXECUTION = "false"
      • Enable only if you trust the environment and need Python tool execution.
    3. ALLOW_SHELL_EXECUTION = "false"
      • Keep false unless shell execution is explicitly required and trusted.
    4. PERSIST_TOOLS = "true"
      • Set to "true" to save custom tools between server restarts.
    5. TOOLS_DB_PATH = "./tools.json"
      • Path where the tools DB is stored. You may change to an absolute container path such as "/app/data/tools.json" if you mount a host volume.
  4. Security reminder (enter these carefully): because this server allows dynamic code execution, do not enable ALLOW_PYTHON_EXECUTION or ALLOW_SHELL_EXECUTION unless you control and trust the deployment environment.
  5. If PERSIST_TOOLS = "true", ensure the TOOLS_DB_PATH points to a persistent/mounted directory:
    1. If using a host volume, mount e.g. host ./data to container /app/data and set TOOLS_DB_PATH = "/app/data/tools.json".
    2. Make sure the directory exists and has write permissions for the MCP process.
  6. Verify boolean values are entered as the string "true" or "false" in the FastMCP interface (the README examples use string values).
  7. Save the connection configuration and complete the Install/Deploy action in FastMCP.
  8. After install, confirm the server started with the chosen env values by checking the FastMCP logs or server status in the FastMCP interface. If you enabled persistence, verify that the tools file (e.g. /app/data/tools.json) is created and writable.

Quick Start

View on GitHub

More for AI and Machine Learning

View All →

More for Developer Tools

View All →

Similar MCP Servers

Report Issue

Thank you! Your issue report has been submitted successfully.