All insights

Copilot Agents · 5 min read

Microsoft's Agent Governance Toolkit: Three Questions to Ask of Any Agent

By James Wilkinson 8 August 2026

Microsoft's open-source Agent Governance Toolkit enforces agent rules in code, not prompts. Why that matters for firms that will never install it.

TL;DR
  • Microsoft open-sourced the Agent Governance Toolkit in April 2026. It sits between an autonomous agent and its tools, checking every action against written policy in ordinary code, giving each agent its own identity and writing a tamper-evident audit log. Most firms will never install it, and it is still worth understanding.
  • Its premise is the important part: an agent's instructions are a request, not a control. Rules that matter get enforced where the agent touches systems, so a denied action is impossible rather than discouraged.
  • It gives every firm three questions to ask of any agent, managed or custom built: what exactly can it do in the systems it can reach, which agent did what, and where is the record that proves it.

Microsoft publishes a lot of developer tooling that businesses can safely ignore. The Agent Governance Toolkit, open-sourced in April 2026, is not quite in that category. Most firms reading this will never install it, and it is still worth ten minutes, because it says out loud what the industry has learned about agents doing real work: instructions are not enforcement.

What it is

The toolkit is a governance layer that sits between an autonomous agent and everything the agent can touch. Every tool call, message and delegation the agent attempts is intercepted in ordinary application code and checked against written policy before it happens. Each agent carries its own identity, so actions are attributable to a specific agent rather than a shared account. And every decision lands in a tamper-evident audit log: the request, the policy in force and any approval given.

It ships for Python, TypeScript, .NET, Go and Rust, plugs into the main agent frameworks including Microsoft’s own Agent Framework and Semantic Kernel, includes a security gateway for MCP tools and maps its controls to OWASP’s agentic AI risk list, NIST’s AI risk framework, the EU AI Act and SOC 2. It is MIT licensed, and Microsoft has said it intends to hand the project to a foundation.

None of that is aimed at a managing partner. The premise behind it is.

The idea that matters

The project’s documentation is blunt about why it exists. Telling an agent its rules in a prompt is, in Microsoft’s words, “a polite request to a stochastic system”. An agent will usually follow its instructions. Usually is not a control. Capable models can be argued, tricked or confused out of their instructions, which is why the toolkit refuses to rely on them: policy is enforced in deterministic code at the point where the agent touches a tool, so a denied action is not discouraged, it is impossible.

That distinction, between asking and enforcing, is the single most useful thing a non-technical leader can take from this project. The instructions you write for an agent shape its behaviour. The permissions and policies around it are what you can actually rely on. Good governance needs both, and it never confuses one for the other.

The three questions

The toolkit frames the problem as three questions, and they translate directly into plain English any firm can use.

1. What exactly can it do? Access to a system is not permission for every operation in it. An agent that reads matters in your practice management system should not be able to delete them, and standard access controls rarely draw the line that finely. The question to ask is not “what can it reach” but “what can it do once it has reached it, and what stops it doing more”.

2. Which agent did what? One agent using your systems is traceable. Several agents sharing a service account are not, and “the integration did it” is no answer during incident response or a client complaint. Every agent needs an identity of its own, so its actions can be separated from yours and from each other’s.

3. Where is the record? When an auditor, a regulator or an unhappy client asks what happened, the answer needs to be a record, not a recollection: what was requested, what policy applied, who approved it. If the honest answer is that no such record exists, the agent is not ready for work that matters.

If your agents live in Microsoft’s platforms

Most firms will meet agents through Microsoft 365 Copilot and Copilot Studio, not through a Python SDK, and on the managed platforms this class of machinery is part of what you are paying for. Agents inherit your existing Microsoft 365 permissions rather than acquiring their own, tasks run in a sandbox governed by the platform and activity lands in the compliance tooling your tenant already has. The governance discipline for client data still applies, because a control you have not configured protects nobody, but the enforcement layer itself comes with the platform.

The toolkit is Microsoft showing its working: the same discipline its own platforms apply, published in the open for everyone building outside them.

If you go custom

The custom-build route is where the three questions become procurement questions. If a partner builds you an agent outside the managed platforms, ask where policy is enforced other than in the instructions, how each agent is identified and where the audit trail lives. You do not need to name the toolkit. You need answers that sound like it. A builder whose full answer is “the prompt tells it not to” has written a request, not a control.

Where human review fits

Enforcement in code does not replace human review points. The two do different jobs. Code draws the hard lines, the things an agent may never do regardless of how convincingly it is asked. Review points cover judgement, the actions that are allowed but should be seen by a person before they stand. The toolkit treats an approval step as part of the policy itself, which matches how we think agents should run everywhere: assisting people, with agreed places where a person checks, never unattended by design.

What to do with this

For most firms there is nothing to install. There is something to write: the policy, in plain English, before any agent exists. What may this agent read. What may it write. What must a person approve. That one page is useful on every platform, managed or custom, and it turns the three questions from an audit surprise into a design input.

Governance is a stage of the agent journey, not an afterthought at the end of it. If agents are on your roadmap and the policy page does not exist yet, our AI governance planning service starts exactly there, or book a consultation and we will write the first draft of those three answers with you.

Sources checked

Last checked: 8 August 2026.

Related reading

More on copilot agents

Copilot Agents Are Copilot Agents Secure? Where Your Data Goes The long answer on Copilot agent security: where answers come from, what stays inside the Microsoft 365 service boundary and what your admins control. Copilot Agents The Copilot Agents Included with a Premium Licence: Researcher, Analyst, Facilitator and More A Microsoft 365 Copilot licence now includes a roster of agents, Researcher, Analyst, Facilitator, app agents and agent building. What each does and what Basic gets. Copilot Agents Microsoft Scout and Autopilots: What Autonomous Agents Mean for Professional Services Microsoft Scout is its first autonomous assistant and the start of a new category called Autopilots. A practical, sceptical guide for UK professional services. Copilot Agents Microsoft Copilot Agents: How AI Moves from Answers to Business Tasks Microsoft Copilot agents move AI from answering questions to supporting repeatable business tasks. A guide to good use cases, source content and governance. Copilot Governance Can You Use Microsoft Copilot with Client Data? A Practical Governance Guide Can you use Microsoft Copilot with client data? Yes, but only inside clear governance, approved tools and a review process built around risk levels. Automation Strategy Copilot, Power Automate or Custom Build? Choosing the Right Automation Route When should you choose Copilot and Power Automate over custom development or an agent? A practical framework for Microsoft 365 automation decisions. Automation Copilot Studio's New Engine: What the GitHub Copilot Harness Changes and What It Costs Copilot Studio's default is now the GitHub Copilot harness, the engine behind Copilot Cowork. What changes, what the credits cover and the checks before you build. Capability map The Agent Journey Five stages from using your first Copilot agent to running a team of agents, with two doors at every stage. Agents Microsoft Copilot agents Plain-English guidance on where Microsoft Copilot agents fit, how to govern them and when to build. Next step Book a free consultation Share where you are now and what you want Microsoft 365 to help with next.

Common questions

Questions about Agent Governance Toolkit

What is Microsoft's Agent Governance Toolkit?
An open-source toolkit, released under the MIT licence in April 2026, that adds a governance layer around autonomous AI agents. It intercepts every tool call an agent attempts and checks it against written policy before it happens, attaches a distinct identity to each agent and records a tamper-evident audit log. It ships for Python, TypeScript, .NET, Go and Rust, works with the main agent frameworks including Microsoft's own Agent Framework and Semantic Kernel, and maps its controls to OWASP's agentic AI risk list, NIST's AI risk framework, the EU AI Act and SOC 2.
Do we need it if our agents run in Copilot Studio or Microsoft 365 Copilot?
No. The toolkit is for teams building agents in code, outside the managed platforms. Agents in Microsoft 365 Copilot and Copilot Studio get this class of control from the platform, which is part of what the licence and credits pay for. The toolkit's three questions are still worth asking of any agent, because they tell you what good governance looks like wherever the agent runs.
What questions should we ask about any AI agent?
Three. What exactly can it do in the systems it can reach, since access to a system is not permission for every operation in it. Which agent did what, since a shared service account cannot answer that. And where is the record that proves what happened, including the policy in force and any approval given. Ask them of a managed platform and of any partner building custom.
Is the Agent Governance Toolkit free?
Yes. It is open source under the MIT licence on GitHub, and Microsoft has said it intends to move the project to a foundation for community governance. It is a developer toolkit rather than a product, so for most businesses its value is the checklist it represents, not the software itself.