Knowledge base
Your policies are the source of truth
Upload PDF, DOCX, TXT, Markdown or CSV files, paste FAQ entries, or import a public URL. Documents are chunked and indexed per workspace, and every AI answer cites the pieces it used.
- File, pasted text and URL import
- Per-workspace index — nothing is shared between tenants
- Chunk-level citations shown on drafts and chatbot answers
- Re-index on demand, with status and error reporting
- Uploads validated by extension and magic bytes, stored outside the web root
Semantic when you configure it, honest when you do not
With an embeddings key, chunks are indexed with a model embedding. Without one, the system falls back to a local hashed vector and labels it as such in the database and the interface — it is usable, but it is not semantic search.
Measured storage
Knowledge-base storage is metered per workspace and enforced server-side against your plan limit.
Imports are treated as untrusted
Imported URLs are fetched server-side with private network ranges blocked, and extracted text is never executed or rendered as HTML into the app.
Ready to give your team time back?
Set up a workspace in minutes. No credit card required for the trial.
14-day trial No card required 9 interface languages Human approval in the loop