Open Source
Explore the latest AI open-source projects from GitHub and HuggingFace.
Explore the latest AI open-source projects from GitHub and HuggingFace.
AstrBot is an open-source agentic instant messaging chatbot infrastructure that integrates over 10 major IM platforms, multiple LLM providers, and nearly 800 community plugins into a single unified framework. With 16,600 GitHub stars and active development since December 2022, it has grown into one of the most versatile chatbot deployment platforms available. ## What Makes AstrBot Unique Unlike single-platform chatbot frameworks, AstrBot is designed from the ground up for cross-platform deployment. A single AstrBot instance can simultaneously serve users on QQ, Telegram, WeChat Work, Feishu, DingTalk, Slack, Discord, and more. This means organizations and individuals can maintain one chatbot configuration that reaches users wherever they communicate, eliminating the need to build and maintain separate bots for each platform. The platform takes an agentic approach to chatbot design, going beyond simple prompt-response patterns. AstrBot supports autonomous agent workflows, MCP (Model Context Protocol) integration, knowledge base retrieval, and multi-modal processing including image understanding and voice interaction. ## Multi-Platform Architecture AstrBot's adapter-based architecture cleanly separates platform-specific communication logic from core AI capabilities. Each supported platform has a dedicated adapter that handles authentication, message formatting, media handling, and platform-specific features like typing indicators and file sharing. Recent updates added automatic typing status display for Telegram and file receiving capabilities for Feishu, demonstrating ongoing platform-specific refinements. The adapter system is extensible, allowing community developers to add support for new platforms. Combined with full internationalization (i18n) support, AstrBot can serve global user bases across different languages and messaging ecosystems. ## LLM Integration AstrBot connects to a broad range of LLM services including OpenAI, Anthropic Claude, Google Gemini, DeepSeek, and local models via Ollama. It also integrates with AI platforms like Dify, Coze, and Aliyun Bailian, giving users flexibility to choose their preferred model provider. The framework handles API key management, rate limiting, and failover between providers automatically. ## Plugin Ecosystem With approximately 800 community-developed plugins available for one-click installation, AstrBot offers extensive customization without writing code. Plugins can add new commands, integrate external services, modify response behavior, or add entirely new capabilities. The plugin system is designed for safety with an Agent Sandbox that provides isolated code execution, preventing malicious plugins from accessing the host system. ## Built-in ChatUI and WebUI AstrBot includes a lightweight ChatUI similar to OpenWebUI for direct browser-based conversations, alongside a comprehensive WebUI for configuration management. The WebUI provides dashboards for monitoring bot performance, managing plugins, configuring LLM connections, and adjusting platform-specific settings. ## Voice and Multi-Modal Capabilities The platform supports voice interaction through Whisper for speech-to-text and multiple TTS options including OpenAI TTS, Edge TTS, and other providers. Multi-modal support enables the bot to process and respond to images, documents, and voice messages, creating richer conversational experiences. ## Deployment and Security AstrBot recommends Docker and Docker Compose for deployment, with additional support for Windows installers and panel-based systems. The project uses pre-commit hooks with ruff for code quality, maintaining high development standards across its 3,900+ commits. Seven active QQ community groups, plus Telegram and Discord channels, provide robust community support.