@FakeMaidenMaker: A lightweight personal AI agent that is small yet highly readable. Existing agent frameworks wrap the core loop in a black box; changing one detail means peeling through layers of abstraction, making even reading a chore, let alone secondary development. I'm recommending an ultra-lightweight open-source AI agent — nanobot. Designed for those who 'want to read and understand…'

Lightweight personal AI agent with small footprint and high readability Existing agent frameworks wrap the core loop in a black box; changing one detail means peeling through layers of abstraction, making even reading a chore, let alone secondary development. I'm recommending an ultra-lightweight open-source AI agent — nanobot. Designed specifically for those who 'want to read and extend', this personal agent is inspired by Claude Code / Codex, with code kept minimal and readable. GitHub: https://github.com/HKUDS/nanobot Its core loop is deliberately written small and clear so anyone can understand the changes. Built-in integration with 10+ chat platforms including Telegram, Discord, Feishu, and WeChat. The WebUI is packaged directly into the wheel; just open a WebSocket to use it, no need to start a separate frontend service. Multi-provider support covers OpenAI, Anthropic, DeepSeek, Gemini, Qwen, as well as local Ollama / vLLM, allowing seamless switching without vendor lock-in. Key features: - Core agent loop is minimal and readable, facilitating research and secondary development - Built-in memory + MCP for long-term memory and tool extensions, ready to use out of the box - /goal command supports maintaining long-term goals across multiple rounds, capable of running long tasks - Integrates with mainstream chat platforms such as Telegram, Discord, Feishu, WeChat - Supports multiple providers including OpenAI, Anthropic, DeepSeek, local Ollama, etc., freely switchable Deploying this project is also simple: `pip install nanobot-ai` then `nanobot onboard`. Suitable for those building a personal AI assistant or wanting to study agent architecture from the source code level.