qmd

What problem does it solve? Finding information across large collections of personal notes, meeting transcripts, and documentation is slow with plain grep and impossible to do semantically without sending data to the cloud. This Skill sets up and operates qmd, a fully local search engine that indexes text files and answers both keyword and conceptual queries on-device. ## Core Features & Use Cases - Three Search Modes: BM25 keyword search for exact terms, semantic vector search for natural-language questions, and hybrid search with LLM reranking for the highest-quality results. - Collection Management: Add directories as named collections, attach context descriptions to improve retrieval, and generate embeddings with a single command. - MCP Integration: Expose qmd as an MCP server (stdio or HTTP daemon) so the agent gets native search tools without loading the Skill each time. - Use Case: You have years of meeting transcripts and project docs. Add them as collections, run qmd embed, then ask "what decisions were made about the database migration" and get reranked, citation-ready results in seconds. ## Quick Start Install qmd with npm, add my notes folder as a collection, generate embeddings, and search for everything about the API redesign.