Compile
Turn policies and regulatory obligations into evidence-linked proposals, then confirm what enters force.
Documentation
Understand the compliance gate. Inspect the proof.
Learn how Avery Rulebook compiles policy, gates live AI decisions and produces auditable evidence. Then connect it through the TypeScript or Python SDK and enforce it in standard agent frameworks.
Reason
Apply the lowest sufficient tier to determine which confirmed rules govern the live context.
Enforce
Compliance-gate answers and actions with allow, deny, approval and obligation decisions.
Prove
Sign every governed decision and record each access for independent audit.
Start here
From written obligation to runtime compliance gate.
Read the system overview first. Then go deeper into the lifecycle, answer ladder, enforcement boundary and audit artifacts.
Foundations3 guides
017 min028 min039 min
How Rulebook works
The public guide to compiling policies and regulatory obligations, compliance-gating AI decisions, enforcing outcomes and proving what happened.
Compile, review and publish
How authoritative source material becomes a signed, immutable rulebook without allowing machine-extracted text to govern by itself.
The answer ladder
Four declared reasoning tiers, a caller-controlled ceiling and one rule: climb only when the tier below cannot answer.
Build4 guides
0910 min1010 min1111 min1212 min
TypeScript SDK
Call Rulebook from Node.js services and place a fail-closed compliance gate in front of tools that create side effects.
Python SDK
Use the synchronous Python client for governed decisions, gateway checks, control-plane workflows and receipt access.
OpenAI Agents SDK
Put Rulebook inside a function tool so the compliance check completes before the OpenAI agent can cause a side effect.
Claude Agent SDK
Map Claude Agent SDK permission checks to Rulebook so governed tools cannot execute before a binding verdict.
Operate3 guides
047 min066 min077 min
Gateway and enforcement
Put policy at the tool-call boundary so an agent checks enterprise authority before a consequential action executes.
Compliance templates
Start from operational control narratives for 20 common frameworks, then adapt, confirm and publish only what applies to your organization.
Interfaces and integrations
Use one compliance decision surface across agent frameworks, enterprise assistants, guardrail layers and traditional applications.
Trust2 guides
058 min088 min
Receipts, ledger and verification
Three trust artifacts answer different questions: what was decided, who accessed what, and whether the proof is intact.
Security and deployment boundaries
Understand what Rulebook protects, what depends on topology, and where enterprise controls remain essential.
Executive briefing
Evaluate one compliance-gated AI workflow end to end.
Bring the governing source material, decision context and action boundary. We will map the Rulebook lifecycle with your compliance, security and AI teams.
Request demo