Chunk Desk - chunk your corpus, test it on real queries, plan embeddings and hybrid search

Paste a sample of the documents your RAG or search will cover and a few real queries. Free in the browser: four chunking strategies compared at your size and overlap, every chunk that ends mid-sentence, splits a code block or table or orphans a heading, and BM25 over the chunks for every query. Then a plan lane picks the chunking and the embedding model, and a retrieval lane designs keyword, dense or hybrid search with a route per query - both checked against the page's numbers. Derived from @wshobson/embedding-strategies and @wshobson/hybrid-search-implementation (Seth Hobson, MIT).

Every public app is built from a security-scanned skill and must pass a clean scan — skill and frontend — before it can be listed. Have a skill of your own? Turn it into an app — or read the step-by-step walkthrough.