Two stores: searchable text and files you keep as files.
Searchable text lives in the database. Photos and PDFs are separate — browse them in admin and use drop-off links.
Knowledge → Index → Retrieve → Agent → Action
How data is stored
Knowledge search is core. Files are separate, with size limits.
knowledge search database
Chunks, embeddings and organization id live in PostgreSQL. Redis and RabbitMQ are for when the action runs, not the knowledge corpus.
File service
The file service stores objects. Attachments from email can land as stored files with a checksum.
Drop links
Anonymous drop links are a files capability, separate from website chat and knowledge search ingest.
No sensitive credentials in graphs
Connection strings and API keys stay in secure credential store references, not in datastore rows of a automated process JSON.
Ready to move beyond the chatbot?
Build AI that understands your business, works with your knowledge, follows your processes and can actually finish the job. You decide which actions are allowed, when a person must step in and how much AI usage your organization can consume.