Search markdown knowledge bases using BM25, semantic vector, and hypothetical document modes.
npx skills add https://github.com/rijnb/knowledge-base-wiki --skill qmd-rijnb
Please help me install this Agent Skill. Skill: qmd Source: https://github.com/rijnb/knowledge-base-wiki/tree/main/.agents/skills/qmd Command: npx skills add https://github.com/rijnb/knowledge-base-wiki --skill qmd-rijnb
Manually sifting through large volumes of markdown notes, knowledge bases, and documentation to find specific information is time-consuming and inefficient, especially when you do not remember exact keywords or phrasing for the content you need.
Use the qmd skill to search your markdown knowledge base for information about how the rate limiter handles burst traffic.
High-intent search queries and answers about installing and using this skill.
To search markdown notes without exact keywords, you can use semantic vector or hypothetical document search modes. These modes match the contextual meaning of your query rather than relying on strict lexical matches to locate relevant information.
The best way to find specific information in a large technical documentation wiki is using a dedicated markdown search tool. Collection filtering allows you to narrow the search scope to specific documentation sets, delivering accurate results faster than manual browsing.
Yes, you can use BM25 lexical search for your personal knowledge base. This search type is supported alongside semantic vector and hypothetical document modes, allowing you to retrieve notes based on exact term frequency and keyword matching.
Hypothetical document search works for retrieving research notes by generating a mock document based on your vague query and using it to find matching content. This method helps locate relevant notes when you only have a general question about the topic.
No, you do not need to manually browse folders to locate content in a team wiki. You can integrate a markdown search server via CLI or MCP to instantly query the knowledge base and retrieve context-aware results without file browsing.
danielcopperMerge, deduplicate, and index Markdown memory files in .claude/memory directories.
SlevoDevEnable semantic search over documents using AgentDB's vector store.
ccfSelect and optimize embedding models for semantic search and RAG applications.
celticht32Design and troubleshoot Couchbase full text and vector search indexes.
