Onyx
Connect users and agents to distributed workplace knowledge through search and chat.
Official product pageAvery Rulebook for Onyx
Onyx connects AI chat to company documents, apps and people. Rulebook adds confirmed rules, deterministic decisions and action receipts when answers carry operational consequence.
The operating reality
The runtime compliance gate
Rulebook evaluates live context against the confirmed policy and regulatory rules that apply, then returns a decision the agent must follow.
Actor, purpose, data, jurisdiction and proposed action
Allow, deny, obligations or named human review
Evidence, rulebook version and signed decision trace
Architecture boundary
Connect users and agents to distributed workplace knowledge through search and chat.
Official product pageTurn selected knowledge into a published, testable decision artifact with explicit authority.
Keep Onyx for knowledge access and conversation. Add Rulebook where enterprise rules must be applied consistently and proved later.
Why the gap persists
Retrieved text can conflict across sources.
Prompt instructions are hard to audit as policy artifacts.
Chat citations do not enforce tool authorization.
Rulebook in the workflow
Onyx retrieves the relevant source material and context.
Rulebook selects the applicable confirmed rule and returns a structured decision.
The agent proceeds, escalates or stops based on the decision.
Potential outcomes
Questions leaders ask
Rulebook supports enterprise deployment patterns and standard interfaces. The final topology should match your security and operations requirements.
No. You can compile only the authoritative materials that should become executable policy.
Executive briefing
Bring one consequential agent workflow. We will map the governing policies and regulations, runtime gate, human authority and auditable evidence path with your team.
Request a briefing