API Reference agentsea_ingest v#0.1.0

Document ingestion. chunk_documents/2 turns documents into chunk messages (the unit the AgentSea.Ingest.Pipeline Broadway topology embeds and stores).

Splits text into overlapping word windows. Overlap preserves context across chunk boundaries (so a fact split between two chunks still embeds coherently in at least one).

A Broadway pipeline that embeds chunk messages and upserts them into a vector store. Concurrency, batching, backpressure, and retries are Broadway settings — there is no hand-rolled scheduler (this is the design's "EvaluationPipeline parallelism bug is structurally impossible" point).