AI Meta MCP Server
AI Meta MCP 服务器
动态 MCP 服务器,允许 AI 模型通过元函数架构创建并执行自定义工具。该服务器提供了一种机制,允许 AI 通过在运行时定义自定义函数来扩展自身功能。
特征
动态工具创建:AI 可以通过自定义实现来定义新工具
多运行环境:支持 JavaScript、Python 和 Shell 执行
沙盒安全:工具在隔离的沙盒中运行以确保安全
持久性:在会话之间存储和加载自定义工具定义
灵活的工具注册表:管理、列出、更新和删除自定义工具
人工审批流程:工具创建和执行需要明确的人工审批
Related MCP server: Code Executor MCP Server
安全注意事项
⚠️警告:此服务器允许动态代码执行。请谨慎使用,并仅在受信任的环境中使用。
所有代码都在沙盒环境中执行
工具创建和执行需要人工审核
可通过环境变量配置工具执行权限
所有操作的审计日志
安装
npm install ai-meta-mcp-server用法
运行服务器
npx ai-meta-mcp-server配置
环境变量:
ALLOW_JS_EXECUTION:启用 JavaScript 执行(默认值:true)ALLOW_PYTHON_EXECUTION:启用 Python 执行(默认值:false)ALLOW_SHELL_EXECUTION:启用 Shell 执行(默认值:false)PERSIST_TOOLS:在会话之间保存工具(默认值:true)TOOLS_DB_PATH:存储工具数据库的路径(默认值:“./tools.json”)
使用 Claude Desktop 运行
将其添加到您的claude_desktop_config.json中:
{
"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"
}
}
}
}工具创建示例
在 Claude Desktop 中,您可以创建一个这样的新工具:
Can you create a tool called "calculate_compound_interest" that computes compound interest given principal, rate, time, and compounding frequency?Claude 将使用define_function元工具来创建您的新工具,该工具可立即使用。
建筑学
该服务器实现了模型上下文协议(MCP),并提供了一个元工具架构,使得人工智能驱动的功能在安全边界内注册和执行成为可能。
执照
麻省理工学院
Available Tools
5 toolsdefine_functionC
Create a new custom MCP function that the AI can use
Inputs: name (required), description (required), parameters_schema (required), implementation_code (required), +1 more.
delete_functionB
Delete a custom MCP function
Inputs: name (required).
get_function_detailsB
Get details of a custom MCP function
Inputs: name (required).
list_functionsB
List all custom MCP functions
Inputs: no named parameters.
update_functionC
Update an existing custom MCP function
Inputs: name (required), description, parameters_schema, implementation_code, +1 more.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 5 tools
Each tool has a clearly distinct purpose targeting different operations on custom MCP functions: define (create), delete, get details, list all, and update. There is no overlap or ambiguity between these CRUD operations.
All tools follow a consistent verb_noun pattern (e.g., define_function, delete_function, get_function_details, list_functions, update_function). The naming is uniform and predictable throughout the set.
With 5 tools, this server is well-scoped for managing custom MCP functions. Each tool serves a clear and necessary purpose, covering the essential operations without being excessive or insufficient.
The tool set provides complete CRUD coverage for the domain of custom MCP functions: define (create), list, get details, update, and delete. There are no obvious gaps, and agents can handle the full lifecycle of functions.
Maintenance
Related MCP Connectors
Runtime permission, approval, and audit layer for AI agent tool execution.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
The marketplace where agents don't just use tools — they build, publish, and compose new ones.
One place to build, share, and govern the skills and tools your AI agents use at work.
Related MCP Servers
- FlicenseCqualityDmaintenanceEnables on-demand generation of TypeScript tools using an LLM with human-in-the-loop approval and safe sandboxed execution. It allows users to create and persist custom functionality through natural language requests.81-
- AlicenseNot gradedqualityCmaintenanceProvides sandboxed code execution for AI agents with support for Python, JavaScript, and shell commands. Includes comprehensive safety features like destructive pattern blocking, timeout protection, and restricted file access for secure production use.35 PyPIMIT
- AlicenseAqualityCmaintenanceEnables dynamic creation and execution of custom tools/functions in multiple programming languages at runtime, exposing them to MCP clients like Claude.410 npm45MIT
- AlicenseAqualityDmaintenanceEnables AI assistants to securely execute Python and JavaScript code in sandboxed environments, with file management and package installation.7MIT