Search local Markdown collections using BM25 and semantic similarity.
npx skills add https://github.com/ramirlm/skillshare-ramir --skill qmd-ramirlm
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/ramirlm/skillshare-ramir/tree/main/qmd-skill Command: npx skills add https://github.com/ramirlm/skillshare-ramir --skill qmd-ramirlm
This Skill addresses the challenge of quickly finding information within large collections of local Markdown files, offering both keyword and semantic search capabilities.
Use the qmd skill to search your notes for "project planning".
High-intent search queries and answers about installing and using this skill.
To search markdown notes locally, you can use a hybrid approach combining BM25 keyword search for exact matches and local LLMs for semantic similarity retrieval. This enables fast information extraction from indexed collections of documents.
Yes, you need Bun installed along with specific system prerequisites to set up local markdown search. These dependencies are required for collection indexing and generating local embeddings for optimal retrieval performance.
BM25 keyword search provides fast exact-match retrieval for markdown documents, while semantic search uses local LLMs for slower, nuanced similarity queries. Hybrid search combines both mechanisms to find related content efficiently.
Yes, you can retrieve specific markdown documents by path or ID. The skill indexes your local markdown collection, allowing direct document retrieval alongside its hybrid keyword and semantic search capabilities.
Semantic search across markdown files using local LLMs is slower than keyword-based BM25 retrieval. It requires generating local embeddings and indexing collections beforehand, which demands specific system prerequisites and setup time.
cdeistopenedIndex and search local Markdown collections with BM25 and vector embeddings.
SigmacodeatSaves notes, files, and piped input into a searchable knowledge database.
giosuetedeschi-specDesign and optimize Retrieval-Augmented Generation pipelines for improved context retrieval.
JuanMS20Index Markdown and text files for local hybrid search with BM25, vectors, and reranking.
