ACL-aware search
Keyword + vector retrieval that only ever returns what the asking user is permitted to see, filtered against cached entitlement snapshots from the source systems.
On-prem · ACL-aware · Agent-ready
Premora ingests from the systems your team already uses, preserves the source permissions through the entire pipeline, and surfaces governed, cited context through search, a deterministic wiki, and a coding-agent skill. The tool that doesn’t guess.
Four ways in
Premora ingests once, preserves ACLs from the source system, then makes the same permissioned knowledge available to humans and agents alike.
Keyword + vector retrieval that only ever returns what the asking user is permitted to see, filtered against cached entitlement snapshots from the source systems.
Raw source content materialized into structured wiki pages, chunks, a knowledge graph, and system pages — with full source lineage on every fact.
A conversational research surface over search, read-only SQL, and deep web research. Durable findings get promoted into the shared wiki under review.
A Claude Code / Codex skill that feeds grounded, cited internal context into AI-assisted coding — so an agent stays anchored to what your team actually knows.
The tool that doesn’t guess
Generic assistants hallucinate confident answers with no source. Premora returns grounded answers with traceable citations and freshness — and only ever surfaces what the asking user is permitted to see. An engineer asks about an incident, a system, or a decision, and gets a sourced answer in under 30 seconds.
Connectors
Every connector projects the source system’s permissions into Premora, so a result is only ever visible to someone already entitled to it upstream.
Premora installs from signed, versioned artifacts — connected, from your own registry, or fully air-gapped from offline media. Start with the docs.