@sailorpepe
基本信息
Zero-trust, local-first FastMCP core for The Undesirables. 35+ offline AI agent tools: TCG card grading, conformal risk forecasts (Safe-Hold/Momentum grades), on-chain soul personalities, x402-payable oracle, TTS + 3D forging. No telemetry. BSL 1.1.
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"undesirables": {
"command": "python3",
"args": [
"-m",
"server"
],
"env": {}
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
Undesirables MCP Server is a Python-based server that provides 35+ local compute tools for AI agents with zero cloud dependency. It integrates with Undesirables NFTs to turn them into MCP-compatible AI agents, enabling vision AI card grading, conformal risk forecasting, AI music generation, local image generation, web search, and more. It is designed for developers and AI agents using MCP clients like Claude Desktop or Cursor.
Install via pip install undesirables-mcp-server or clone the repository and set up a virtual environment. Boot the server with python server.py --workspace "/path/to/soul_folder" (requires Ollama running and Python 3.10+). Connect to a chat front-end by configuring an MCP client (e.g., Claude Desktop) with the server command and workspace path.
You need Python 3.10 or higher, Ollama installed and running (the app must be double-clicked and active), and a virtual environment for the server. Optional: a Hugging Face read token for local image generation on Apple Silicon.
Configure your MCP client (e.g., Claude Desktop) with the server command `python server.py --workspace "/path/to/your_soul_folder"` and the `cwd` pointing to the server directory. Restart the client to see the tools.
Make sure Ollama is physically running (the llama icon should appear in your menu bar or system tray). The server requires Ollama to be active for local inference.
The server is designed to work with a soul workspace (downloaded from the Undesirables website). Without a soul folder, the agent may not have a personality or memory; but the core tools (e.g., web search, image generation) may still be available if you provide a workspace path.
The project is licensed under BSL-1.1. The README includes a link to the LICENSE file for details.
Hindsight: Agent Memory That Learns
A good sequence diagram can solve almost any communication problem in engineering. Volley lets your agent write them as plain text, with the real request and response payloads on every arrow. It checks its own syntax wi
A curated list of references for MLOps
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
Connect your AI assistant to your own YOUB endurance training account. Read recorded activities with full analysis, planned sessions, race goals, wellness (HRV, sleep, resting HR) and your athletic profile. Create or mov
评论