agentscope-ai — 13 Open-Source AI Agent Skills

agentscope-ai is the author of 13 open-source AI agent skills and MCP servers spanning mcp-server, codex-skill, agent-tool, claude-skill, with a combined 83,940+ GitHub stars. On AgentSkillsHub each is quality-scored (avg 66/100) and security-graded — 12 verified safe.

Top skills by agentscope-ai

• QwenPaw — Your Personal AI Assistant; easy to install, deploy on your own machine or on the cloud; supports multiple chat apps with easily extensible (35.2k★ · quality 67/100 · security: safe)
• agentscope — Build and run agents you can see, understand and trust. (32.2k★ · quality 70/100 · security: safe)
• AgentTeams — An open-source Collaborative Multi-Agent OS for transparent, human-in-the-loop task coordination via Matrix rooms. (5.7k★ · quality 71/100 · security: caution)
• HiClaw — An open-source Collaborative Multi-Agent OS for transparent, human-in-the-loop task coordination via Matrix rooms. (4.9k★ · quality 70/100 · security: safe)
• ReMe — ReMe: Memory Management Kit for Agents - Remember Me, Refine Me. (3.5k★ · quality 69/100 · security: safe)
• agentscope-runtime — A production-ready runtime framework for agent apps with secure tool sandboxing, Agent-as-a-Service APIs, scalable deployment, full-stack ob (847★ · quality 66/100 · security: safe)
• OpenJudge — OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards (829★ · quality 60/100 · security: safe)
• agentscope-samples — A collection of ready-to-use Python sample agents built with AgentScope and AgentScope Runtime, covering use cases from CLI tools to full-st (336★ · quality 58/100 · security: safe)

View agentscope-ai on GitHub · Explore audited agent skills