Index and search local documents using BM25, vector embeddings, and reranking.
npx skills add https://github.com/hcnimi/clawdis --skill qmd-hcnimi
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/hcnimi/clawdis/tree/main/skills/qmd Command: npx skills add https://github.com/hcnimi/clawdis --skill qmd-hcnimi
Traditional local search either relies on keyword matching, which misses semantic intent, or uses expensive embedding pipelines that are impractical for everyday workflows. qmd combines BM25 indexing with vector-based retrieval and a reranking step to deliver fast, accurate results over your local documents.
Index your local documents with qmd and perform a search to retrieve relevant results.
High-intent search queries and answers about installing and using this skill.
You can index and search local documents using a combination of BM25 lexical retrieval and vector-based semantic search to capture both exact keywords and underlying intent across personal knowledge bases and local corpora.
Applying a reranking step after initial BM25 indexing and vector retrieval improves top-result quality, ensuring the most accurate and relevant matches are surfaced for your local code notes.
Yes, the tool supports an MCP mode integration, allowing you to incorporate fast local document indexing and retrieval into advanced, modular query workflows via a Node-based installation.
You can use provided CLI commands to index, update, check status, and search across your local corpora, managing the entire retrieval pipeline directly from the command line.
No, combining BM25 indexing with vector-based retrieval and a reranking step delivers fast, accurate results over local documents without requiring expensive, impractical embedding pipelines.
Traditional keyword search relies solely on lexical matching, missing semantic intent, whereas combining BM25 with vector-based semantic search ensures meaningful matches across your local files.
lpjhelderInstall and configure kronon-forge with Node.js and npm.
LauraFlorentinRetrieves relevant documents from a knowledge base to ground responses.
Yangel-hideSearch local knowledge stores with hybrid BM25, vector, and LLM reranking.
jacardlPerform fast vector similarity search with filtering via REST and gRPC APIs.
