Agents et multi-agents

LangBot

langbot-app/langbot

An open-source, production-grade platform for building and deploying AI-powered instant messaging bots across multiple platforms.

★ 17,2KÉtoiles
⑂ 1,5KForks
124Problèmes ouverts
PythonLangue
Apache-2.0Licence
Q@project.QualityScoreScore éditorial

Captures d'écran du projet

Capture d’écran de LangBot Capture d’écran de LangBot Capture d’écran de LangBot Capture d’écran de LangBot Capture d’écran de LangBot Capture d’écran de LangBot

Vue d’ensemble

LangBot connects Large Language Models (LLMs) to various chat platforms, enabling the creation of intelligent agents capable of multi-turn dialogues, task execution, and workflow integration. It features a web management panel for configuration and monitoring, a multi-pipeline architecture for handling different bot scenarios, and extensive plugin and integration support.

Fonctionnalités clés

  • Universal IM Platform Support (Discord, Telegram, Slack, LINE, QQ, WeChat, WeCom, Lark, DingTalk, KOOK, Matrix, Email)
  • AI Conversations & Agents with multi-turn dialogues, tool calling, multi-modal support, and streaming output
  • Built-in RAG (knowledge base) with deep integration to Dify, Coze, n8n, Langflow, Deerflow, and Weknora
  • Plugin Ecosystem with hundreds of plugins, event-driven architecture, and MCP protocol support
  • Web Management Panel for configuration, management, and monitoring without YAML editing
  • Multi-Pipeline Architecture for managing different bots and scenarios
  • Production-Ready features including access control, rate limiting, sensitive word filtering, and exception handling
  • Agent-friendly design with a built-in MCP Server, in-repo skills directory, and AGENTS.md

Prérequis, installation et démarrage rapide

One-Line Launch: `uvx langbot` (requires uv). Docker Compose: clone the repository, navigate to the `LangBot/docker` directory, and run `docker compose --profile all up -d`. Cloud deployment options are also available via Zeabur and Railway.

Utilisation

After installation, access the web management panel at http://localhost:5300 to configure IM platform connections, select LLM providers, set up pipelines, and manage plugins. A live demo is available at https://demo.langbot.dev/ with email `demo@langbot.app` and password `langbot123456`.

Compatibilité des modèles et cas d’usage

Supports OpenAI, Anthropic, DeepSeek, Google Gemini, xAI, Moonshot, Zhipu AI, Ollama, LM Studio, and various gateways like SiliconFlow, Aliyun Bailian, Volc Engine Ark, ModelScope, GiteeAI, 302.AI, and Qiniu. Integrates with LLMOps platforms like Dify.

Licence et notes sur les risques

Apache-2.0

Editorial verification 2026-08-09: repository URL, owner, description, license and repository statistics were reviewed. License metadata: Apache-2.0. README was fetched for the channel draft; re-check repository dependencies, releases and model terms before production use.

Publication et maintenance

Not stated in the repository metadata

Firecrawl

firecrawl/firecrawl

★ 161,1KTypeScript

LangChain

langchain-ai/langchain

★ 143,6KPython

RAGFlow

infiniflow/ragflow

★ 86,7KGo