# zhayujie/CowAgent

Open-source super AI assistant & Agent Harness. Plans tasks, runs tools and skills, self-evolves with memory and knowledge. Multi-model, multi-channel. Lightweight, extensible, one-line install. (formerly chatgpt-on-wechat)

Repository: https://github.com/zhayujie/CowAgent
Canonical: https://ross.abutalabs.com/products/cowagent
Homepage: https://cowagent.ai
Language: Python
License: MIT
License Family: permissive
Topics: openai, ai, llm, ai-agent, deepseek, mcp, skills, claude, openclaw, chatgpt-on-wechat, harness, ai-agents, claude-code, codex, cowagent, multi-agent
Last push: 2026-08-26T09:10:15+00:00

## Health v2 (maintenance only)
Score: 99/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 98, longevity 100
- inputs: {"age_days": 1487, "days_push": 7, "days_rel": 12, "gap_med": 12.0, "n_releases_24m": 23}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 46692, forks 10328 (observed 2026-08-28T04:12:11.106585+00:00)

## What it is
CowAgent (formerly chatgpt-on-wechat) is an open-source AI assistant and Agent Harness that plans tasks, runs tools and skills, and self-evolves with long-term memory and a personal knowledge base. It supports pluggable LLM providers and deploys across web and major IM channels like WeChat, Feishu, and DingTalk, running 24/7 on a laptop or server.

## Use cases
- run a personal AI assistant on wechat
- build an ai agent that plans and executes tasks
- chatgpt bot for dingtalk or feishu
- self-hosted ai assistant with long-term memory
- create custom ai skills from natural language
- connect llm to mcp tools and browser automation
- personal knowledge base with knowledge graph

## When to choose
- you want a multi-channel AI assistant across WeChat, Feishu, DingTalk, QQ, and web
- you need task planning with tool/skill execution and MCP integration
- you want self-evolving memory and an auto-curated knowledge base
- you want to swap between many LLM providers with one click

## When to avoid
- you need only a simple stateless chatbot without memory or planning
- you require a fully managed cloud service with no self-hosting
- you need a lightweight library to embed in an existing app rather than a standalone assistant

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, chatbot, llm-inference, mcp, rag, workflow-automation, chat-interface, scheduling
- domain: artificial-intelligence, large-language-models, chatbots, self-hosted, developer-tools
- platform: python, windows, self-hosted, cross-platform, cli
- tags: agent-harness, skills-system, multi-channel, long-term-memory, knowledge-base, task-planning, multi-model, personal-assistant, im-integration, self-evolution, ai-agents, retrieval-augmented-generation, automation, linux, macos, docker, wechat, web-server

## Member repositories
- zhayujie/CowAgent (main) score 99

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:12:11.106585+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:21:36.419898+00:00, confidence not recorded.
  - readme: https://github.com/zhayujie/CowAgent (fetched 2026-08-28T04:12:11.106585+00:00, sha c9f63cccbe85)
  - homepage: https://cowagent.ai (fetched 2026-08-29T07:45:04.829438+00:00, sha 38576bd88aae)
  - site_page: https://docs.cowagent.ai (fetched 2026-08-29T07:45:04.835533+00:00, sha 2623220cfa0d)
  - site_page: https://docs.cowagent.ai/skills/index (fetched 2026-08-29T07:45:04.837566+00:00, sha bb3abf717e73)
- Data as of 2026-08-30T08:39:29.467469+00:00.