Microsoft Agent Governance Toolkit

🌍 [English](/README.md) | [日本語](./docs/i18n/README.ja.md) | [简体中文](./docs/i18n/README.zh-CN.md)

Welcome to Agent Governance Toolkit!

[](https://github.com/microsoft/agent-governance-toolkit/actions/workflows/ci.yml) [](LICENSE) [](https://python.org) [](packages/agent-mesh/sdks/typescript/) [](https://www.nuget.org/packages/Microsoft.AgentGovernance) [](packages/agent-mesh/sdks/rust/agentmesh/) [](packages/agent-mesh/sdks/go/) [](docs/OWASP-COMPLIANCE.md) [](https://www.bestpractices.dev/projects/12085) [](https://scorecard.dev/viewer/?uri=github.com/microsoft/agent-governance-toolkit) [](https://deepwiki.com/microsoft/agent-governance-toolkit)

[!IMPORTANT] **Public Preview** — All packages published from this repository are **Microsoft-signed public preview releases**. They are production-quality but may have breaking changes before GA. For feedback, please [open a GitHub issue](https://github.com/microsoft/agent-governance-toolkit/issues).

**What this toolkit is:** Runtime governance infrastructure — deterministic policy enforcement, zero-trust identity, execution sandboxing, and reliability engineering that sits between your agent framework and the actions agents take.

**What this toolkit is not:** This is not a model safety or prompt guardrails tool. It does not filter LLM inputs/outputs or perform content moderation. It governs *agent actions* (tool calls, resource access, inter-agent communication) at the application layer. For model-level safety, see [Azure AI Content Safety](https://learn.microsoft.com/azure/ai-services/content-safety/).

Runtime governance for AI agents — the only toolkit covering all **10 OWASP Agentic risks** with **9,500+ tests**. Governs what agents *do*, not just what they say — deterministic policy enforcement, zero-trust identity, execution sandboxing, and SRE — **Python · TypeScript · .NET · Rust · Go**

...