Skip to content
Brutor AI Docs
The AI System assurance loop An AI System's approved operating envelope is declared, enforced on every runtime call to models, MCP tools, A2A agents and skills, and continuously proven from the evidence each run leaves behind — which feeds back into the envelope. 1 · DECLARE Approved operating envelope 2 · ENFORCE Every call, at runtime 3 · PROVE Continuous evidence LLMs · MCP · A2A · skills Proof feeds back into what the system may do next

The assurance control plane for AI Systems

Know what your AI Systems are allowed to do. Control them at runtime. Continuously prove that they remain within their approved operating envelope.
Quickstart — run BrutorYour first governed LLM call

Pick your path

I'm building an agent or app

Point your OpenAI or Anthropic SDK at Brutor with a one-line base-URL change, call MCP tools through a governed endpoint, or build a portal on the Portal API.

I'm running AI Systems in production

Declare each agent as an AI System and Brutor watches it: liveness, drift, contract conformance and a health verdict that never reports green on absent evidence.

I'm installing and operating Brutor

Run the full platform with Docker Compose in about ten minutes, then take the governance and observability tours.

Your AI System, end to end

What is an AI System?

The unit you govern and assure: the models, MCP tools, skills and agent identities an agent uses, bound together, with every run attributed back to it.

Develop → Deploy → Govern → Observe → Assure

One journey from first prototype to production evidence: build on governed rails, promote through the lifecycle gate, let the run ledger prove how it behaves.

What you build with Brutor

Agents

LangGraph, Claude Agent SDK or plain HTTP — agents inherit guardrails, budgets and audit automatically.

Custom portals & chat clients

End-user auth, workspaces, model discovery, streaming chat, tool approvals and notifications — bring your own UI.

MCP servers & tools

Register a server once and every governed client gets it with access control and tool-call policies.

Setup & provisioning scripts

Everything the Admin UI does is a REST API. Load users, groups, models, guardrails and budgets from an idempotent script.

The platform

Governance

Resource groups, guardrails, argument & semantic policies, budgets, quotas, approvals and compliance tagging.

AI System Assurance

The run ledger, liveness, drift, contracts, replay and the health verdict.

Observability

Mission Control dashboards, proxy & usage logs, audit trail, Prometheus metrics and Grafana.

API Reference

Runtime & Portal APIs on the Core Proxy (:8100), Admin API on the Control Plane (:5050).

Architecture

How the Rust core proxy, Python control plane, Admin UI and User Portal fit together.

Compliance

Sealed action records, witnessed tenant logs and obligation evidence for the EU AI Act, ISO 42001, SOC 2, HIPAA, GDPR and more.