Index and search local markdown and text documents with hybrid retrieval.
npx skills add https://github.com/peteromallet/megaplan --skill qmd-peteromallet
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/peteromallet/megaplan/tree/main/megaplan/agent/optional-skills/research/qmd Command: npx skills add https://github.com/peteromallet/megaplan --skill qmd-peteromallet
Local, on-device search across markdown notes, meeting transcripts, documentation, and other text files using a hybrid retrieval approach (BM25, vector embeddings, and LLM reranking) to deliver fast, private results.
Install qmd, index your collections, and start querying your local data.
High-intent search queries and answers about installing and using this skill.
You can search markdown notes and transcripts locally using hybrid retrieval, which combines BM25 keyword matching, vector embeddings, and LLM reranking. This on-device approach keeps your personal knowledge base private while delivering fast results.
Yes, you need Node.js version 22 or higher, along with SQLite extension support on macOS and the qmd package, to index local text documents and perform on-device retrieval across your personal collections.
Yes, this skill supports MCP integration to expose local search tools directly to Hermes Agent, allowing the agent to query your indexed markdown notes and documentation on-device.
Hybrid retrieval combines BM25 keyword scoring, vector embeddings for semantic similarity, and LLM reranking to find relevant results. This multi-layered approach improves search accuracy across personal knowledge bases and project documentation.
On-device indexing and search provides a private alternative to cloud-based tools. By processing markdown and text files locally with hybrid retrieval, your documentation remains completely on your machine without external data transmission.
ldzhhxxIndex and search markdown documents with BM25, vector search, and LLM reranking.
steipeteIndex local files and search them with BM25, vector, and hybrid queries.
jason660519Search local notes and transcripts with hybrid BM25, vector, and reranking retrieval.
RheasilviaIntegrate Pinecone vector databases for retrieval-augmented generation and long-term memory.
