Documentation

Operate · Guide 07

Interfaces and integrations

Use one compliance decision surface across agent frameworks, enterprise assistants, guardrail layers and traditional applications.

7 min read

Pre-release documentation. This guide describes the current product design and evaluation build. Packaging, availability and commercial terms may change before release.

01

One semantic core, several interfaces

The same rulebook and decision semantics can be exposed through REST, MCP, the command line and language SDKs. Integration kits map platform lifecycle events to those interfaces without copying policy logic into the kit.

  • REST for applications and services
  • MCP for agent tool discovery and calls
  • CLI for operators, review workflows and automation
  • TypeScript, Python and Rust clients for product teams
02

Where Rulebook connects

  • Agent frameworks: LangChain, LangGraph, OpenAI Agents, Claude Agent SDK and Gemini agents
  • Enterprise agents: Salesforce Agentforce and ServiceNow AI Agents
  • Knowledge systems: Glean and Onyx
  • Guardrail and gateway products: NeMo Guardrails, Guardrails AI, Bifrost, Lakera Guard, WitnessAI and AWS Bedrock Guardrails
  • Agent gateways and self-hosted environments such as OpenClaw
03

Integration rule

Kits contain mapping logic, not business policy. Platform churn is absorbed at the adapter boundary while rule semantics remain in the published Rulepack.

Executive briefing

Apply the architecture to one consequential workflow.

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