Platform
Knowledge Corpora
Ground your agents in company knowledge. Ingest documents, sync SharePoint or sitemaps, and connect corpora so every answer is backed by your sources.
Q4 Report.pdf
Onboarding Guide.docx
API Reference.md
HR Policies.pdf
1,247 docs · 84,392 chunks indexed
How it works
Ejento's Knowledge Corpora transform your existing documents, wikis, and data sources into a retrieval layer your agents draw from in real time. Upload PDFs, Word documents, and Markdown files, or sync directly from SharePoint, Confluence, or web sitemaps. Documents are chunked, embedded, and indexed so agents can surface precise, cited answers rather than hallucinated responses. Access controls ensure each agent only retrieves from corpora it is permitted to use.
Key capabilities
- Ingest PDF, DOCX, MD, HTML, and plain text
- Native sync with SharePoint, Confluence, and sitemaps
- Automatic chunking, embedding, and vector indexing
- Per-agent corpus access controls
- Citation-backed answers where agents show their sources
- Incremental re-indexing when documents change