Retrieve and answer questions from indexed knowledge base collections using hybrid BM25 and vector search.
npx skills add https://github.com/MuduiClaw/ClawKing --skill kb-rag
Please help me install this Agent Skill. Skill: kb-rag Source: https://github.com/MuduiClaw/ClawKing/tree/main/workspace/skills/kb-rag Command: npx skills add https://github.com/MuduiClaw/ClawKing --skill kb-rag
This Skill allows you to efficiently retrieve information and answer questions from a vast and growing knowledge base, ensuring you can access and utilize accumulated information effectively.
Use the kb-rag skill to find information about 'OpenClaw 2026.2.13 updates'.
High-intent search queries and answers about installing and using this skill.
Hybrid search for knowledge base retrieval combines BM25 keyword matching with vector similarity to find relevant context, which an LLM then synthesizes into a precise answer with sources.
You can ingest new content into a knowledge base by adding documents to indexed collections, making them immediately available for future retrieval and question answering.
Yes, you can search across multiple indexed collections including blog translations, project documentation, skill documentation, and conversation captures to retrieve targeted information.
Yes, manual search operations and content ingestion for knowledge base retrieval require qmd to interface with the indexed collections and execute queries.
The best way to query conversation captures and project documentation is to ask natural language questions, leveraging hybrid search to get synthesized answers with cited sources.
sky770825Search local Markdown collections using BM25 keyword and semantic embeddings.
thecloudtipsSynthesize sourced answers with inline citations and confidence ratings.
steipeteIndex local files and search them with BM25, vector, and hybrid queries.
Division6066Combine vector similarity with keyword matching for hybrid search retrieval.
