English | 简体中文

🔥 The agent skill ecosystem is exploding—over 90,000+skills are now publicly available.

But with so many options, how do you find the right skills for your task? And when one skill isn’t enough, how do you compose and orchestrate multiple skills into a working pipeline?

AgentSkillOS is the operating system for agent skills—helping you discover, compose, and run skill pipelines end-to-end.

• 🔍 Skill Search & Discovery — Creatively discover task-relevant skills with a skill tree that organizes skills into a hierarchy based on their capabilities.
• 🔗 Skill Orchestration — Compose and orchestrate multiple skills into a single workflow with a directed acyclic graph, automatically managing execution order, dependencies, and data flow across steps.
• 🖥️ GUI (Human-in-the-Loop) — A built-in GUI enables human intervention at every step, making workflows controllable, auditable, and easy to steer.
• ⭐ High-Quality Skill Pool — A curated collection of high-quality skills, selected based on Claude's implementation, GitHub stars, and download volume.
• 📊 Observability & Debugging — Trace each step with logs and metadata to debug faster and iterate on workflows with confidence.
• 🧩 Extensible Skill Registry — Easily plug in new skills, bring your own skills via a flexible registry.

👉 View detailed workflows on Landing Page →

📊 Check out the comparison report: AgentSkillOS vs. without skills →

Task: I'm a short-video creator. Generate a cat meme video featuring a boss (Angry Cat) questioning an employee (Sad Cat) about work progress, with witty responses. Use video.mp4 (green-screen footage) and background.jpg. Requirements: remove green-screen, maintain aspect ratio, add "Boss"/"Employee" labels, sync subtitles with meowing, and create humorous dialogue with viral potential.

Generated Video:

Task: I'm a product designer planning a knowledge management software. Research products like Notion and Confluence, then create a visual design style report (report.docx) with screenshots. Based on the analysis, generate three design concept images (fusion_design_1/2/3.png) that synthesize their design characteristics.

Generated Design Concepts:

Generated Design Style Report:

Task: I'm a front-end developer. Users reported a bug when accessing my login page on mobile devices. Please identify and fix the bug, then generate a bug report with before/after screenshots highlighting the issue (For demonstration purposes, the bug screenshot is shown below).

Generated bug fix screenshots:

Generated Bug Report:

Task: As a PhD student, I've completed a research paper (Avengers.pdf) and want to promote it on social media platforms. Help me create promotional materials that effectively present my research findings to a broader audience.

Generated Promotional Materials:

Xiaohongshu (RedNote) Post:

Other Social Media Content:

Scientific Slides:

• Skill tree construction: Organizes over 90,000+ skills into a capability tree, providing structured, coarse-to-fine access for efficient and creative skill discovery.
• Skill retrieval: Automatically selects a task-relevant subset of usable skills given a user’s request.
• Skill orchestration: Composes the selected skills into a coordinated plan (e.g., a DAG-based workflow) to solve tasks beyond the reach of any single skill. Note that we also support a freestyle mode (i.e., Claude Code).

Left: Pure semantic retrieval prioritizes texutal similarity, often missing skills that look unrelated in embedding space but are crucial for actually solving the task—leading to narrow, myopic skill usage.

Right: Our LLM + Skill Tree navigates the capability hierarchy to surface non-obvious but functionally relevant skills, enabling broader, more creative, and more effective skill composition.

git clone https://github.com/ynulihao/AgentSkillOS.git
cd AgentSkillOS
pip install -e .
cp .env.example .env  # Edit with your API keys
python run.py --port 8765

# .env
LLM_MODEL=openai/anthropic/claude-opus-4.5
LLM_BASE_URL=https://openrouter.ai/api/v1
LLM_API_KEY=your-key

EMBEDDING_MODEL=openai/text-embedding-3-large
EMBEDDING_BASE_URL=https://api.openai.com/v1
EMBEDDING_API_KEY=your-key

• Interactive Agent Execution
• Plan Refinement
• Auto Skill Import
• Dependency Detection
• History Management
• Recipe Generation & Storage
• Multi-CLI Support (Codex, Gemini CLI, Cursor)

If you find AgentSKillOS useful, consider citing our paper:

@article{li2026agentskillos,
  title={Leveraging, Managing, and Scaling the Agent Skill Ecosystem},
  author={Li, Hao and Mu, Chunjiang and Chen, Jianhao and Ren, Siyue and Cui, Zhiyao and Zhang, Yiqun and Bai, Lei and Hu, Shuyue},
  journal={Preprint},
  year={2026}
}