agentskit.js
Data layerRAG

RAG

Plug-and-play retrieval. Chunk, embed, search, rerank, loaders.

Core pipeline

  • createRAG({ embed, store }) — one-liner ingest + retrieve + search.
  • chunkText({ chunkSize, chunkOverlap, split }) — standalone splitter.

Reranking + hybrid

  • createRerankedRetriever — plug in Cohere Rerank, BGE, or the built-in BM25.
  • createHybridRetriever — vector + BM25 blend with weighted normalization.
  • Recipe: RAG reranking.

Document loaders

  • loadUrl, loadGitHubFile, loadGitHubTree, loadNotionPage, loadConfluencePage, loadGoogleDriveFile, loadPdf (BYO parser).
  • Recipe: Document loaders.

Per-loader + per-reranker deep dives land in step 4 of the docs IA rollout.

✎ Edit this page on GitHub·Found a problem? Open an issue →·How to contribute →

On this page