# microsoft/agent-governance-toolkit

AI Agent Governance Toolkit — Policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents. Covers 10/10 OWASP Agentic Top 10.

Repository: https://github.com/microsoft/agent-governance-toolkit
Canonical: https://ross.abutalabs.com/products/agent-governance-toolkit
Language: Python
License: MIT
License Family: permissive
Topics: agent-framework, ai-agents, ai-safety, compliance, governance, microsoft, owasp, policy-engine, python, security, trust, zero-trust
Last push: 2026-08-25T18:09:18+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 87, longevity 13
- inputs: {"age_days": 184, "days_push": 8, "days_rel": 85, "gap_med": 7, "n_releases_24m": 16}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 6128, forks 1078 (observed 2026-08-28T04:09:36.236667+00:00)

## What it is
A Microsoft open-source toolkit providing policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering for autonomous AI agents, covering the OWASP Agentic Top 10. It ships as a unified installer with SDKs for Python, JavaScript, and .NET and works with any agent framework.

## Use cases
- enforce runtime policies on what autonomous agents are allowed to do
- add zero-trust identity and authorization to AI agent tool calls
- sandbox agent execution to prevent dangerous actions like dropping database tables
- audit and monitor agent behavior for compliance
- secure multi-agent systems against OWASP Agentic threats
- add SRE-style reliability engineering to production AI agents

## When to choose
- you are deploying autonomous AI agents to production and need governance controls
- you need to satisfy security or compliance requirements for agentic systems
- you want framework-agnostic policy enforcement and sandboxing for agent tool use

## When to avoid
- you only need a simple agent framework without security or governance requirements
- you need a fully GA-stable product, since this is a public preview with possible breaking changes
- your project is not built around autonomous agents or LLM tool use

## Facets
- artifact type: library
- maturity: experimental
- function: security, auth, authorization, agent-framework, middleware, monitoring, logging
- domain: security, large-language-models, developer-tools, legal
- platform: python, cross-platform
- tags: ai-safety, governance, policy-engine, zero-trust, sandboxing, owasp-agentic-top-10, compliance, audit, multi-agent, sre, public-preview, ai-agents, nodejs

## Member repositories
- microsoft/agent-governance-toolkit (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:09:36.236667+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T17:47:58.955269+00:00, confidence not recorded.
  - readme: https://github.com/microsoft/agent-governance-toolkit (fetched 2026-08-28T04:09:36.236667+00:00, sha c5ace24b87fe)
  - registry_pypi: https://pypi.org/pypi/agent-governance-toolkit/json (fetched 2026-08-29T08:44:29.007810+00:00, sha 9cd6e87d4949)
- Data as of 2026-08-30T08:39:29.467469+00:00.