QMD Local Search
Performs lightning-fast local semantic and keyword searches across markdown files, documentation, and notes using local LLM reranking.
QMD Local Search brings advanced document retrieval capabilities to Claude Code by leveraging the qmd CLI. It replaces slow, traditional file discovery tools like 'find' with a sophisticated hybrid engine that combines BM25 full-text search, vector semantic search, and local LLM reranking. This skill is ideal for developers and technical writers managing large documentation sets or personal knowledge bases, allowing for precise content discovery and context gathering without ever sending data to external APIs.
Características Principales
01Hybrid search engine combining BM25, semantic vectors, and LLM reranking
02Local model execution for embeddings and reranking with no API keys required
03Collection-based indexing to scope searches to specific projects or folders
04Advanced file retrieval with support for specific line ranges and byte limits
05High-speed file discovery that outperforms standard filesystem search utilities
0616 GitHub stars
Casos de Uso
01Searching through large documentation repositories to find specific implementation patterns
02Replacing standard 'find' commands to locate configuration files in complex directory structures
03Extracting relevant snippets from local notes to provide context for coding tasks