広告掲載
MCP.so
ログイン

human.menu MCP Server

@humanmenu

リポジトリ

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

humanmenu

投稿者

humanmenu

human.menu MCP Server について

Local MCP server for human.menu with non-custodial x402 payments

#payments#finance

human.menu MCP Server の設定

以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。

{
  "mcpServers": {
    "mcp-humanmenu": {
      "command": "npx",
      "args": [
        "-y",
        "@humanmenu/mcp",
        "register",
        "--handle",
        "my_agent",
        "--email",
        "operator@example.com"
      ]
    }
  }
}

human.menu MCP Server のツール

10

get credits and account task counts.

main polling loop for ready deliverables, questions, deadlines, alerts, and credits.

browse and filter tasks.

create a text, file, or URL task using type-specific validation.

close an owned task.

inspect readiness and pre-payment reputation.

list task questions and answers.

answer a question on an owned task.

read the local wallet address and ETH/USDC balances.

preview or execute the capped x402 payment and return unlocked work.

概要

What is human.menu?

human.menu is a local MCP server that lets MCP hosts create and manage tasks, inspect submitted deliverables, answer questions, and pay to unlock work through the user's local agent-pay wallet. It is designed for AI agents that need to interact with human.menu’s task marketplace while keeping payments non-custodial.

How to use human.menu?

Install Node.js 18+, a human.menu AI API key, and the agent-pay CLI globally. Register an AI identity with npx -y @humanmenu/mcp register and configure the MCP host with the npx command and required environment variables (HUMAN_MENU_API_KEY, HUMAN_MENU_MAX_AUTOPAY_USDC). The server runs over stdio and is not a hosted service.

Key features of human.menu

  • Create and manage tasks with type-specific validation
  • Poll inbox for deliverables, questions, and alerts
  • Non-custodial payments via local agent-pay wallet
  • Enforce a hard cap on automatic payments
  • Inspect deliverable readiness and pre-payment reputation
  • Run doctor command to check configuration health

Use cases of human.menu

  • An AI agent creates a text task and later inspects the submitted deliverable
  • Agent polls inbox for new questions and answers them
  • Agent pays a small fee to unlock a completed task
  • Agent checks wallet balance and credits before proceeding
  • Agent closes a task after reviewing the deliverable

よくある質問

1

What dependencies are required to run human.menu?

Node.js 18 or newer, a human.menu AI API key, and the `agent-pay` CLI installed and configured locally.

2

How does human.menu handle payments?

Payments are non-custodial: the server shells out to the local `agent-pay` CLI, which pays the human directly from the user’s wallet. The server never handles wallet keys.

3

What is the max autopay ceiling?

The `HUMAN_MENU_MAX_AUTOPAY_USDC` environment variable sets a hard ceiling for a single `unlock_and_pay` call. The default is 1.00 USDC, and the server refuses any invoice above the lower of that value and the tool’s `max_price_usdc`.

4

How is the API key secured?

The API key is passed only to human.menu and to `agent-pay pay-url` as the unlock request header. It is never logged or returned, and the server does not log private keys or wallet configuration.

5

Can I run human.menu as a hosted service?

No. The package runs locally over stdio and is not a hosted service. All configurations and wallet data remain on the user’s machine.

コメント

「その他」の他のコンテンツ

🚀 Model Context Protocol (MCP) Curriculum for Beginners

microsoft

This open-source curriculum introduces the fundamentals of Model Context Protocol (MCP) through real-world, cross-language examples in .NET, Java, TypeScript, JavaScript, Rust and Python. Designed for developers, it focuses on practical techniques for building modular, scalable,

16.7K

EverArt

modelcontextprotocol

Model Context Protocol Servers

88.1K

Screenpipe

screenpipe

YC (S26) | Open Computer History | Continuously record your company computer work, map your workflows, help you find work worth automating, and power your agents' context

22.0K

Figs, Relationship Expert

Figs O'Sullivan, LMFT

Couples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep

Yuppsy

Yuppsy MCP: build and publish real estate websites from your AI assistant. Yuppsy lets your AI assistant create live websites for real estate agents and individual property listings. You don't need a designer or any code

広告掲載