菜单
Route RAG vector database decisions across Qdrant setup, production operations, and datastore selection by data type.
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
Use this skill when building, debugging, or improving Retrieval-Augmented Generation systems, including chunking, vector database selection, hybrid search, reranking, multimodal RAG, code documentation RAG, retrieval latency, and production RAG architecture.
Chunk nested documents into parent-child levels so retrieval can move from broad sections to fine-grained passages.
Use semantic boundaries and embedding similarity to chunk text for higher-relevance retrieval.
Route RAG chunking decisions across semantic, hierarchical, sliding-window, contextual-header, and framework-selection strategies.
Use overlapping windows to preserve context across chunk boundaries while controlling retrieval size.
Reduce retrieval latency with caching, batching, and index-level optimization.
| name | vector-databases |
| title | Vector Databases |
| description | Route RAG vector database decisions across Qdrant setup, production operations, and datastore selection by data type. |
| category | vector-databases |
| tags | ["vector-database","qdrant","metadata","rag"] |
| allowed-tools | ["Read","Grep","Glob"] |
Use this parent skill when the main RAG problem is choosing, configuring, or operating the vector storage layer. Route to the child skill that matches setup, production, or datastore-selection needs.
Vector database mistakes often show up as slow queries, weak filtering, poor metadata modeling, or expensive production operations. RAG systems need a storage layer that matches the data type, scale, and query pattern.
Separate text, code, multimodal, and metadata-heavy retrieval requirements.
Choose collection layout, vector fields, payload indexes, and metadata conventions.
Validate latency, backup strategy, migration path, monitoring, and failure recovery.
