Enable hybrid BM25 and vector search across local markdown and JSONL files.
npx skills add https://github.com/lev-os/lev-content --skill qmd-lev-os
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/lev-os/lev-content/tree/main/sources/skills/lev-find-qmd Command: npx skills add https://github.com/lev-os/lev-content --skill qmd-lev-os
This Skill provides fast, full-text and semantic search capabilities across local markdown and JSONL files, including conversation histories, making it easy to find specific information without relying on external services.
claude-sessions, claudesp-sessions, and clawdbot-sessions by running a single command.Search for "authentication jwt middleware" across your local session files.
High-intent search queries and answers about installing and using this skill.
You can search local markdown and JSONL files using hybrid search, which combines BM25 full-text matching with vector similarity to retrieve relevant information from indexed local data stores.
Hybrid search for local text retrieval combines BM25 keyword matching with vector embedding similarity, ensuring you find both exact phrases and semantically related content across your local collections.
Yes, you can index and manage multiple collections of text files, allowing you to search across various Claude variant conversation histories simultaneously using a single query.
Yes, it provides an MCP server for agent integration, enabling automated systems to perform hybrid search and retrieve information directly from your locally indexed markdown and JSONL data stores.
You can update the search index by running collection management commands, which scan your local files and refresh the BM25 and vector embeddings to include newly added markdown and JSONL content.
okbest2024Store and organize user-defined memory in a local ~/memory/ directory.
hxk622Analyze Neuropixels recordings with SpikeInterface and Kilosort4.
samymityRefine retrieved code context through iterative relevance-scored retrieval cycles.
chronista-clubEnable semantic retrieval over high-dimensional embeddings using Qdrant.