---
id: agent-governance-toolkit
version: "4.1.0"
license: MIT
license_treatment: permissive
maintenance: active
---

# agent_governance_toolkit — Public Preview — Unified installer and runtime policy enforcement for the Agent Governance Toolkit

License: permissive · Maintenance: active · Downloads: 130.6K/mo

## What it is and what it does

Agent Governance Toolkit is a meta-package that provides a unified installation for governance infrastructure across four layers: a policy engine for deterministic action evaluation, trust infrastructure for agent identity and secure communication, an execution supervisor for resource and context control, and reliability engineering tools for SLO enforcement and incident response. It solves the problem of deploying AI agents to production without governance by bundling version-compatible components that work together out of the box. The package is currently in public preview, meaning APIs may change before general availability, and it is designed to integrate with popular agent frameworks.

Use it for:

- Enforce capability-based access control for agents calling external tools or databases in regulated environments.
- Establish zero-trust identity and encrypted communication between multiple agents in a distributed system.
- Prevent runaway agents from consuming unbounded resources through execution rings and resource limits.
- Generate immutable audit trails and compliance attestations for agent decisions in regulated organizations.
- Monitor agent health and automatically trigger incident response when reliability thresholds are breached.

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Unified installer that bundles policy enforcement, agent identity and trust infrastructure, execution supervision, and reliability monitoring into a single governance stack for production AI agents.

Yes, if you are building production AI agents and need integrated governance across policy, trust, execution control, and reliability. The unified install eliminates version conflicts and ensures component compatibility. However, the package is in public preview with potential API changes; evaluate your tolerance for breaking changes before general availability. No known security vulnerabilities and low install friction support adoption.

## Install

pip install agent-governance-toolkit
uv add agent-governance-toolkit
poetry add agent-governance-toolkit

## Installing agent_governance_toolkit

Before you install: Low friction install with three common dependencies (pydantic, pyyaml, click). Active maintenance with recent commits; package is in public preview with potential API changes before general availability.

License in practice: MIT license permits unrestricted use, modification, and distribution with minimal attribution requirements, suitable for both open-source and commercial projects.

Quickstart:

pip install agent-governance-toolkit

from agent_governance_toolkit import GovernanceVerifier

verifier = GovernanceVerifier()
attestation = verifier.verify()
print(f"Grade: {attestation.compliance_grade()}")

Requires Python 3.9 or later; package is in public preview with APIs subject to change before general availability.

Verify before relying:

- Which of the four component layers (kernel, trust mesh, runtime, SRE) are included in base install versus optional extras.
- Specific OWASP compliance controls and regulatory frameworks (GDPR, HIPAA, SOX) actually implemented.
- Performance characteristics and scalability limits for governed agent execution.
- Actual compatibility with LangChain, CrewAI, and AutoGen beyond installation examples.

## Package facts

- License: MIT (permissive)
- Python support: supports_current
- Install friction: low
- Maintenance: active
- Downloads: 130.6K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

ai agent governance, agent policy enforcement, multi-agent security, agent runtime control, ai compliance and audit, agent-governance, policy-enforcement, multi-agent-security

[View on SkillFed](https://skillfed.io/packages/agent-governance-toolkit) · [View on PyPI](https://pypi.org/project/agent-governance-toolkit/)