Product
Complete context for your people and your AI agents.
What troveGEN is
troveGEN turns your documents into something your apps and AI agents can search accurately. You send it files, web pages or feeds. It reads them, including scans, makes sense of their structure, protects sensitive details and indexes them. Your app then asks questions through one API and gets back the right passages, with their sources, from only what each person is allowed to see.
The platform
Seven capabilities, one governed layer.
Each page below lists exactly what it does today.
INGESTION
Documents in, structure kept, ready for search.
Read scans, tables, books and web pages properly.
Explore ingestion →
SEARCH
Find the right passage, for the right person.
Keyword and meaning search with ranking and filters.
Explore search →
ASK AND CHAT
Answers with sources, or an honest “not found”.
Cited answers, exact numbers, team chat.
Explore ask and chat →
CONNECTORS
Keep your index in step with where documents live.
Keep documents in step with where they live.
Explore connectors →
SECURITY AND CONTROL
Governed access to documents, for people and for AI agents.
Permissions, masking, roles, audit, SSO.
Explore security and control →
QUALITY AND INSIGHTS
Measure your search, don’t assume it.
Measure search, watch usage, prove improvements.
Explore quality and insights →
CONTEXT FOR AI AGENTS
Complete context for your people and your AI agents.
MCP server, SDK and API for agents.
Explore context for ai agents →
How it fits together
From a file to a governed answer.
What goes in
- PDFs & scansTyped or photographed pages
- Spreadsheets & tablesRows stay searchable
- Web pages & feedsCrawled or synced
- Email & slidesWord, PowerPoint, .eml
What troveGEN does
- 1ReadsScans, tables and layouts, in many languages
- 2ProtectsSensitive details replaced before indexing
- 3IndexesFully managed, or inside your own database
- 4FindsKeyword and meaning search, ranked and filtered
What comes out
- The right passageWith its source page and section
- Exact numbersTotals computed from the table, not guessed
- Only what is allowedEach person sees just their own documents
- An honest "not found"Questions the documents cannot answer are refused
Questions people ask
The short version.
What does troveGEN do?
troveGEN turns your documents into something your apps and AI agents can search accurately. You send it files, web pages or feeds. It reads them, including scans, makes sense of their structure, protects sensitive details and indexes them. Your app then asks questions through one API and gets back the right passages, with their sources, from only what each person is allowed to see.
Is troveGEN a vector database?
No. A vector database stores and searches vectors. troveGEN is everything around it: reading and structuring documents, protecting sensitive data, embedding, hybrid search, ranking, permissions, answers and evaluation. It writes into a vector database it hosts for you, or into one you already run.
Who is it for?
Teams building AI products that need document search without building the plumbing, regulated organisations that need control over where data lives, and teams who want colleagues to ask questions of their documents and get cited answers.