Search local Markdown collections using BM25 and optional embeddings.
npx skills add https://github.com/vanducng/skills --skill qmd-vanducng
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/vanducng/skills/tree/main/skills/qmd Command: npx skills add https://github.com/vanducng/skills --skill qmd-vanducng
Local Markdown search with BM25 and optional vector embeddings for fast, on-device retrieval across personal notes, design-system catalogs, and knowledge bases.
Install qmd globally, create or point to a Markdown collection, and run a quick search to retrieve the top matching documents.
High-intent search queries and answers about installing and using this skill.
On-device Markdown search indexes local collections and retrieves relevant documents by path, docid, or glob. It operates locally without sending data to external services, ensuring privacy and fast retrieval.
BM25 search ranks Markdown documents based on keyword relevance and term frequency. It is the default retrieval method for quickly locating specific design notes, meeting transcripts, or knowledge base entries.
Yes, you can enable optional semantic search with embeddings for your Markdown knowledge base. This provides vector-based reranking and LLM rerank capabilities when deeper relevance is needed beyond default BM25.
The best way to manage multiple Markdown collections is to point to your directories and run indexing updates. This refreshes the indices as your data changes, keeping document retrieval accurate.
On-device Markdown search operates without sending data to external services. Optional embedding-based vsearch and LLM rerank are supported locally, ensuring retrieval works without relying on an external internet connection.
You should not use default BM25 alone when you need deeper semantic relevance for local Markdown files. In such cases, switch to optional embedding-based vsearch and LLM rerank for better contextual retrieval.
NITISH-gitbitCombine BM25 keyword matching, vector semantic search, and LLM reranking for local Markdown file retrieval.
nirholasSearch local Markdown notes using BM25 keyword and optional vector search.
overviewlabsSearch local Markdown collections with hybrid keyword and vector retrieval.
JuanMS20Index Markdown and text files for local hybrid search with BM25, vectors, and reranking.
