Search local knowledge bases with BM25, vector search, and LLM reranking.
npx skills add https://github.com/aayushsoam/clawbot-agent --skill qmd-aayushsoam
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/aayushsoam/clawbot-agent/tree/main/optional-skills/research/qmd Command: npx skills add https://github.com/aayushsoam/clawbot-agent --skill qmd-aayushsoam
This Skill solves the problem of efficiently searching through personal knowledge bases, notes, documents, and meeting transcripts stored locally. It provides a powerful search engine that combines keyword matching, semantic understanding, and LLM-powered reranking.
To start using qmd, first add your collections with qmd collection add path/to/collection --name collection_name. Then, you can search your knowledge base with qmd search "query".
High-intent search queries and answers about installing and using this skill.
You can search local markdown notes on your device using a hybrid search engine that combines BM25 keyword matching, vector search, and LLM reranking. It operates entirely locally to find precise results from your personal knowledge base.
Yes, running this local knowledge base search requires Node.js and SQLite. It is currently supported exclusively on macOS, relying on these dependencies to process markdown notes and meeting transcripts on your device.
The best way to combine keyword and semantic search for documentation is using a hybrid search approach. It integrates BM25 for keyword matching, vector search for semantic understanding, and LLM reranking to deliver highly precise results.
Yes, you can integrate local search into your workflows using its built-in CLI and MCP support. After adding collections via the CLI, you can query your markdown notes and meeting transcripts directly from other MCP-compatible tools.
You add markdown files to the local search engine by running the command `qmd collection add path/to/collection --name collection_name`. Once indexed, you can search your notes using `qmd search "query"`.
SolomonikVikExtract MacWhisper SQLite transcripts into timestamped Markdown files with speaker labels.
dawsonblockSearch local knowledge bases with hybrid BM25, vector, and LLM reranking.
RasaHQConfigure Rasa assistants with EnterpriseSearchPolicy for vector store knowledge base search.
giuseppe-trisciuoglioIntegrate Qdrant vector database with Java Spring Boot and LangChain4j for embedding storage and similarity search.
