Retrieval-augmented, source-cited, and private to your team.
PDF, DOCX, TXT. We extract, chunk and embed automatically.
Ask follow-up questions. Threaded history is saved per workspace.
Every claim links back to a file and page number you can verify.
Token-by-token responses powered by Lovable AI.
Row-level security. Your documents never leave your workspace.
No API keys to manage. Sign in and start uploading.
Pick a plan that fits your team.
PDFs, Microsoft Word (.docx) and plain text (.txt).
It retrieves the most relevant chunks from your uploaded documents using vector search, then asks an LLM to answer using only that context.
No. If the answer isn't in your documents, it will say so explicitly.
Yes. Documents are scoped per user with row-level security and stored in a private bucket.