
agent-governance-toolkit
Microsoft toolkit for policy enforcement, identity and audit of autonomous agents
- Pricing
- Open source
- Platforms
- CLI, API, self-hosted
- GitHub stars
- 6.6K
- Last commit
- 1 day ago
- Language
- Python
- License
- MIT
About agent-governance-toolkit
agent-governance-toolkit is an open-source AI tool for security. It is available on CLI, API and self-hosted.
The README frames agent safety as three questions: is this action allowed, which agent did it, and can you prove what happened. It argues that prompt-level instructions are not a control surface and that OAuth scopes control which services an agent can reach, not what it does once connected.
The toolkit answers with policy enforcement, agent identity, sandboxing and SRE practices, installed with one pip command and designed to work with any framework. SDKs are published for Python, npm and NuGet, and docs include an OWASP agentic top-10 architecture mapping.
It is a governance library you embed rather than an agent that acts for you, so it is listed as a tool. Licence is MIT.
What it does
- Policy enforcement
- Checks whether an agent action is allowed.
- Agent identity
- Tells apart agents that share infrastructure.
- Tamper-evident audit
- Records which policy was active and why a request was allowed or denied.
- Sandboxing and SRE
- Containment and reliability practices for agents.
- Multi-language SDKs
- Python, npm and NuGet packages.
Good for
- Blocking an agent from running destructive database commands
- Giving each agent in a system its own identity
- Producing audit records for regulators
- Mapping controls to the OWASP agentic top 10
Pricing
Open source
Free, MIT
agent-governance-toolkit FAQ
Which frameworks does it work with?
The README says one pip install works with any framework.
Which languages have SDKs?
Python, JavaScript (npm) and .NET (NuGet).
Who maintains it?
It is hosted under the microsoft GitHub organisation.
Is agent-governance-toolkit free?
Yes. agent-governance-toolkit is open source, so you can use and self-host it for free.
What are the best alternatives to agent-governance-toolkit?
The closest alternatives to agent-governance-toolkit are ECC, firecrawl, ponytail, graphify and orca.
Alternatives to agent-governance-toolkit

Skills, hooks, memory and security layer for coding agent harnesses
- Freemium
- Open-source agent
- 305K GitHub stars
- AI models and LLM tooling

Web search, scraping and page interaction API for AI agents and developers
- Freemium
- Agent framework
- 202K GitHub stars
- Web search and scraping

Ruleset that makes AI coding agents write less code that still works
- Open source
- Open-source agent
- 192K GitHub stars
- AI models and LLM tooling

Code knowledge graph that gives coding assistants shared, source-linked context
- Freemium
- Open-source agent
- 149K GitHub stars
- Knowledge and memory

Desktop environment for running several coding agents in parallel worktrees
- Open source
- Open-source agent
- 127K GitHub stars
- Coding

Agent skill that turns questions or code into validated HTML diagrams and one-page answers
- Open source
- Open-source agent
- 97.2K GitHub stars
- Design and media