AgenticContract

AgenticContract Overview

Policy engine for AI agents — enforceable rules, risk limits, approval gates, obligation tracking, and violation detection in a single .acon file

AgenticContract is a policy engine for AI agents. It models policies, risk limits, approvals, obligations, and violations in a single binary .acon file.

Why AgenticContract

Every AI agent today is ungoverned. Claude can exceed your API budget. GPT can bypass approval gates. Your copilot can violate compliance rules without anyone noticing. Every action happens without oversight.

AgenticContract gives your agent a governance engine — behavior constraints, budget enforcement, approval gates, and breach tracking — all structured, all queryable in microseconds.

Core Capabilities

Capability	What It Does
Policies	Define allow/deny rules scoped to global, session, or task levels
Risk Limits	Enforce rate, budget, and threshold limits with real-time tracking
Approval Gates	Route sensitive actions through configurable approval workflows
Obligations	Track time-bound compliance requirements with deadline enforcement
Violation Detection	Record and query policy breaches with severity and actor attribution
Conditions	Evaluate boolean conditions that gate policy activation

How It Works

Create a .acon contract artifact
Add policies, limits, approval rules, and obligations
Check every agent action against the contract in microseconds
Enforce limits, route approvals, and track violations automatically
Query the full governance state at any time

Install

curl -fsSL https://agentralabs.tech/install/contract | bash

Or via package managers:

cargo install agentic-contract-cli agentic-contract-mcp
pip install agentic-contract
npm install @agenticamem/contract

File Format

AgenticContract uses the .acon binary format — portable, self-contained, and designed for 20-year durability. One file holds your entire governance state: policies, limits, approvals, obligations, violations, and audit history.

Works With

AgenticMemory — memory-aware policy decisions
AgenticVision — visual observation governance
AgenticCodebase — code change approval gates
AgenticIdentity — identity-bound policy enforcement
AgenticTime — temporal obligation tracking