This is your complete reference when configuring GrepAI's vector search for codebases. It covers all the YAML options: embedder setup (Ollama, OpenAI, LM Studio), storage backends (local gob files, Postgres, Qdrant), chunking strategies, and search score boosting. What I appreciate here is the practical use case examples, like how to configure for maximum privacy versus team environments, plus the specific recommendations on chunk sizes and overlap ratios. The boosting configuration is smart too, letting you penalize test files and vendor directories while surfacing actual source code. If you're moving beyond defaults or troubleshooting why search results feel off, this gives you all the knobs to turn.
npx skills add https://github.com/yoanbernabeu/grepai-skills --skill grepai-config-reference