# AI Agent Governance

> Decide what an agent may do, who can approve a change, and what evidence to keep.

Published by the team behind Emisar. These educational examples do not certify a system as safe or compliant.

## Guides
- [What is AI agent governance?](https://aiagentgovernance.dev/text/what-is-ai-agent-governance.md): Understand who should own an AI agent, what it may do, and how to review its actions.
- [An AI agent governance framework for engineering teams](https://aiagentgovernance.dev/text/ai-agent-governance-framework.md): Build a repeatable review process for agent ownership, access, testing and changes.
- [Write an AI governance policy for agents](https://aiagentgovernance.dev/text/ai-agent-governance-policy.md): Use a practical policy structure that connects agent permissions to owners and enforcement.
- [AI agent access control at the action boundary](https://aiagentgovernance.dev/text/ai-agent-access-control.md): Decide which agent may perform which action on which resource, and enforce the decision.
- [Human in the loop AI that people can actually review](https://aiagentgovernance.dev/text/human-in-the-loop-ai.md): Design approval steps that give a person enough context and authority to make a useful decision.
- [How to assess the risk of an AI agent](https://aiagentgovernance.dev/text/ai-agent-risk-assessment.md): Review the harm an agent could cause, the controls that limit it, and the evidence still missing.
- [What belongs in an AI agent audit trail](https://aiagentgovernance.dev/text/ai-agent-audit-trail.md): Record agent requests, access decisions and outcomes without copying secrets into your logs.
- [Least privilege for AI agents](https://aiagentgovernance.dev/text/least-privilege-ai-agents.md): Limit an agent's authority to the actions, resources and time its task actually needs.
- [AI agent identity and delegated access](https://aiagentgovernance.dev/text/ai-agent-identity.md): Keep the agent, the person requesting work and the executing service identifiable throughout a task.
- [AI governance vs AI security](https://aiagentgovernance.dev/text/ai-governance-vs-ai-security.md): Understand how governance decisions and security controls work together in an agent deployment.
- [Production access for AI agents](https://aiagentgovernance.dev/text/production-access-for-ai-agents.md): Give an agent useful operational access with explicit scope, approval rules and recoverable execution.
- [Incident response for an AI agent](https://aiagentgovernance.dev/text/ai-agent-incident-response.md): Contain unexpected agent activity, preserve evidence and restore access only after checking the cause.

## About
- [Ownership and editorial policy](https://aiagentgovernance.dev/about/)
- [Tools we build](https://aiagentgovernance.dev/tools-we-build/)