LangBot
Production-grade platform for building agentic IM bots - 生产级多平台智能机器人开发平台/ Agent、知识库编排、插件系统 / Bots for Discord / Slack / LINE / Telegram / WeChat(企业微信, 企微智能机器人, 公众号) / 飞书 / 钉钉 / QQ / Matrix e.g. Integrated with ChatGPT(GPT), DeepSeek, Dify, n8n, Langflow, Coze, Claude, Gemini, GLM, Ollama, SiliconFlow, Moonshot, openclaw / hermes agent, deerflow
LangBot is an open-source platform that connects AI language models to chat apps like Discord, Telegram, Slack, and WeChat so you can build and deploy conversational AI assistants without writing a server from scratch.
LangBot is an open-source platform for building AI-powered chat bots that live inside instant-messaging apps. The README describes it as a production-grade platform that connects large language models to chat platforms so you can create agents that hold conversations, run tools, and plug into existing workflows. Out of the box it supports Discord, Telegram, Slack, LINE, QQ, WeChat, WeCom (Enterprise WeChat), Lark, DingTalk, KOOK, Satori, Email, and Matrix, which can bridge to networks such as Signal, WhatsApp, and iMessage.
A single LangBot server sits between the chat platforms and an AI model of your choice. It integrates with cloud LLMs including OpenAI, Anthropic, DeepSeek, Google Gemini, xAI, Moonshot, and Zhipu AI; with locally-run models through Ollama and LM Studio; and with orchestration stacks like Dify, Coze, n8n, and Langflow. It supports the MCP protocol for connecting tools to models. Conversations can be multi-turn, multi-modal, and streamed, and there is a built-in retrieval-augmented generation feature for grounding answers in a knowledge base. A browser-based management panel lets you configure pipelines, plugins, access control, rate limits, and sensitive-word filtering without editing config files. Hundreds of plugins extend behavior through an event-driven architecture.
Someone would reach for LangBot to ship an internal company assistant on Slack or Lark, a community helper on Discord or Telegram, or a customer-facing agent on WeChat. Quick-start options include a one-line uvx command, Docker Compose, one-click deploys to Zeabur or Railway, and a hosted Cloud version. The project is written in Python and supports versions 3.10 through 3.13.
Where it fits
- Deploy an internal company assistant on Slack or Lark that answers questions from a knowledge base.
- Build a community helper bot on Discord or Telegram backed by any cloud or local AI model.
- Create a customer-facing agent on WeChat or WeCom that handles common support queries.
- Connect a local Ollama model to multiple chat platforms through a single self-hosted server.