Knowledge Ingestion
Upload PDFs or add URLs. Content is embedded into the knowledge base the voice agents retrieve from.
Upload PDF files
Text is extracted and embedded. Scanned PDFs with no text layer are reported, not silently skipped. Files are sent one at a time so progress is real.
Select PDF files
Add URLs
Each page is fetched, stripped of markup and embedded.
URLs to process
Crawl depth
Depth above 1 crawls same-domain links and can index many pages - it takes longer.
Verify retrieval
Runs the same query the voice agents run, against the same tenant - the honest check that an upload actually landed.
Ingested sources0
| Source | Type | Chars | Status | Actions |
|---|---|---|---|---|
| Nothing ingested for this tenant yet. | ||||