HomeProjectsArticlesPK BoardAbout
Explore Projects

CowAgent

Active
GitHub Python MIT

Description

CowAgent (formerly chatgpt-on-wechat) is a powerful AI assistant framework built on LLMs with autonomous planning, tool use, long-term memory, multi-agent collaboration, and multi-channel integration for WeChat, Feishu, DingTalk, and more.

Key Features

  • Three-tier memory architecture — Context→Daily→Core memory system with automatic Deep Dream distillation and hybrid keyword+vector retrieval
  • Skill system — One-click install from Skill Hub or create custom skills through natural language conversation
  • Multi-channel integration — Supports WeChat, Feishu, DingTalk, WeCom, QQ, Telegram, Slack and 10+ platforms
  • Autonomous task planning — Automatically decomposes complex tasks and executes step-by-step, looping tools until goal is reached
  • Structured knowledge management — Auto-curates knowledge into Markdown Wiki with visual knowledge graph browsing
  • Multi-model routing — Route Claude, GPT, Gemini, DeepSeek, Qwen to different functions with one-click Web console switching

Use Cases

💡 Enterprise WeChat/Feishu smart assistant: Deploy AI assistants on enterprise messaging platforms for automated customer and internal support
💡 Personal knowledge management: Automatically organize personal knowledge bases through conversation records with searchable long-term memory
💡 Automated workflow orchestration: Connect external services via Skill system for cross-platform automated task processing
💡 Multi-channel customer service bots: Provide consistent AI customer service across WeChat, DingTalk, web and other channels simultaneously

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (46.8k stars)
  • • Permissive open-source license (MIT)
  • • Established track record (4 years in production)

Tags

agent-framework multi-agent chatbot wechat llm mcp python

Categories

🤖 Agent Framework 🧠 Multi-Agent

Quick Start

# Linux / macOS one-line install
bash <(curl -fsSL https://cdn.link-ai.tech/code/cow/run.sh)

# Docker deployment
curl -O https://cdn.link-ai.tech/code/cow/docker-compose.yml
docker compose up -d

# Access Web console
# http://localhost:9899

# Service management
cow start | stop | restart
cow skill install <name>
Visit GitHub Visit Website

Related Projects

AG2

4.9k · Python
Active A+

AG2 (formerly AutoGen) is an open-source AgentOS providing a multi-agent conversation framework with flexible agent orchestration, tool integration, and distributed collaboration for building complex multi-agent systems.

multi-agentframeworkagent +3
  • · Multi-agent conversation framework with flexible orchestration patterns including swarms, group chats, and nested chats
  • · Support for multiple LLM providers with configurable API keys and model selection
  • · Human-in-the-loop workflows enabling human oversight and input in agent conversations
View on GitHub

Langroid

4.1k · Python
Active A+

Langroid is a Python multi-agent programming framework that leverages an intuitive Agent-Task-Tool abstraction to help developers build LLM-powered multi-agent applications.

multi-agentllmagent-framework +1
  • · Multi-agent programming: intuitive Agent-Task-Tool abstraction inspired by the Actor Framework
  • · LLM-agnostic: works with practically any LLM, including local models via OpenAI-compatible APIs
  • · MCP tool adapter: convert MCP server tools into Langroid ToolMessage instances
View on GitHub

LazyLLM

3.9k · Python
Active A+

LazyLLM is a lightweight multi-agent LLM application framework offering the easiest way to build multi-agent LLM apps, with built-in RAG, knowledge graph, fine-tuning, and integration with LangChain and LlamaIndex ecosystems.

multi-agentframeworkrag +3
  • · Low-code multi-agent application assembly with built-in data flow and functional modules like Lego blocks
  • · One-click deployment with lightweight gateway for POC and Kubernetes packaging for production
  • · Cross-platform compatibility across bare-metal, Slurm clusters, and public clouds without code changes
View on GitHub

Ouroboros

5.8k · Python
Active A

Ouroboros is a spec-driven multi-agent framework that shifts from traditional prompting to specification-driven development, supporting multi-agent collaboration, MCP tool integration, and automated workflow orchestration for building high-quality agent systems.

multi-agentframeworkagent +4
  • · Spec-first workflow: Socratic interview → immutable seed spec → execution → evaluation
  • · Multi-runtime adapter supporting Claude Code, Codex CLI, OpenCode, Hermes, Gemini, Kiro, Copilot
  • · Three-layer architecture: Kernel (contract engine), Plugins (domain workflows), Shell (TUI cockpit)
View full details → View on GitHub