f4bc0d3/huggingmes-hermes-webui

huggingmes-hermes-webui is a ready-to-deploy package that lets you run a personal AI agent with a chat interface on Hugging Face Spaces at no cost. The author is clear that this is not original work, it combines three existing open-source projects into one easy setup: Hermes Agent by Nous Research (the AI logic layer), Hermes WebUI by @nesquena (the browser chat interface), and HuggingMes by @somratpro (the Hugging Face deployment wrapper). Setup takes around five minutes: duplicate the Hugging Face Space, add three secrets (a password you invent for logging in, a Hugging Face token for saving your data, and a Cloudflare Workers token to keep the Space from sleeping), then add an API key for one of the supported AI providers, OpenAI, Anthropic, or Moonshot. After a first build of around five to eight minutes, a chat interface is available at your Space's URL. All chat sessions, agent memory, workspace files, profiles, and settings are automatically backed up to a private Hugging Face Dataset every ten minutes when a Hugging Face token is configured. If the Space restarts, everything is restored from the latest snapshot. Beyond chat, the deployment exposes an OpenAI-compatible API at /v1/* that other applications can connect to using the same login token. An optional Telegram bridge is available if a Telegram bot token is added. MCP (Model Context Protocol) servers can also be configured to extend the agent's capabilities. Internally, a Node.js router on a single port directs traffic to the Hermes WebUI, dashboard, gateway API, and Telegram webhook as separate subprocesses. The full README is longer than what was provided.