Skip to content
Acme RAG

Documents

The corpus behind every answer, and the access tier that decides who can retrieve from it.

Loading corpus...

Add a document

Uploads need the ingestion pipeline - PDF parsing, BGE-M3 embeddings and pgvector - which the hosted demo does not run. Clone the repo and start the FastAPI stack to ingest your own PDFs; everything else here works against the bundled corpus.

No content available at your access level.

Retrieval you can put in front of an auditor

A multi-agent assistant over a small company knowledge base. Unlike a general chatbot it enforces who may read what, cites the passage behind every claim, and refuses when the evidence is not there. Three things make that true.