菜单
Learn RAG by building real applications. Use when (1) building first RAG from scratch, (2) understanding RAG component design, (3) debugging retrieval problems, (4) optimizing retrieval quality, (5) comparing RAG frameworks.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Complete software development lifecycle from requirements to deployment. Use when (1) starting a new project from scratch, (2) need structured end-to-end development process, (3) require comprehensive documentation and quality gates at each phase.
专业的AI Agent(AI Agents)顾问助手，探索 AI Agent 框架和应用。当用户询问以下问题时使用：(1) 技术选型和对比 (2) 使用指南和最佳实践 (3) 问题诊断和解决 (4) 资源推荐 (5) 常见问题解答
Comprehensive CV learning assistant. Use when studying image processing, object detection, segmentation, or any CV tasks. Helps with algorithm understanding, implementation, and model optimization.
Comprehensive DL learning assistant. Use when studying neural networks, CNN, RNN, LSTM, Transformer, or any DL architectures. Helps with network design, training strategies, debugging, and optimization techniques.
Comprehensive LLM learning assistant. Use when studying transformer architecture, attention mechanisms, pre-training, fine-tuning, or prompt engineering. Helps with understanding LLM principles and practical applications.
Comprehensive ML learning assistant. Use when studying supervised learning, unsupervised learning, regression, classification, clustering, or any ML concepts. Helps with algorithm understanding, implementation guidance, model evaluation, and practical applications.
| name | rag-learning |
| description | Learn RAG by building real applications. Use when (1) building first RAG from scratch, (2) understanding RAG component design, (3) debugging retrieval problems, (4) optimizing retrieval quality, (5) comparing RAG frameworks. |
Build working RAG with LangChain + Chroma in 30 minutes.
For step-by-step: See references/level1-minimal.md
Understand and experiment with each RAG component.
For details: See references/level2-components.md
Handle PDF, web pages, tables. Build personal knowledge base.
For details: See references/level3-documents.md
Implement Hybrid Search, Reranking, better prompts.
For details: See references/level4-optimization.md
Compare frameworks, choose vector DB, add caching.
For details: See references/level5-production.md
| Use Case | Framework | Why |
|---|---|---|
| Quick prototype | LangChain | Most examples, easy start |
| Data-heavy apps | LlamaIndex | Best data connectors |
| No prompting | DSPy | Programmatic LLM |
| Production RAG | RAGFlow | End-to-end solution |
| Problem | Solution |
|---|---|
| Chunk too large/small | Adjust chunk_size (500-1000 tokens) |
| Wrong embedding model | Try domain-specific model |
| Retrieved but wrong answer | Check prompt template |
| No results | Lower similarity threshold |
| Slow retrieval | Add caching, use hybrid search |
