# Create a policy from plain language

Start with a business boundary and turn it into a control the runtime can enforce.

- Category: Policy Builder
- Last updated: 2026-06-19
- URL: https://help.irongorilla.ai/articles/create-a-policy-from-plain-language

## Write the boundary first

Good policies start with the business boundary: routine work, risky work, sensitive data, and the moments where a person needs to stay involved. The builder can help translate that boundary into a control, but the owner still reviews the final behavior with [policy simulation](/articles/simulate-a-policy-before-launch).

![Iron Gorilla policy editor with plain-language policy intent, generated rule, and simulation output.](https://help.irongorilla.ai/help/screenshots/policy-simulation.png)

_Policy simulation lets teams see how a rule behaves before it reaches live work._

## Creating the draft policy

Create policies from Governance by describing the boundary in plain language. Review the generated rule, the scope where it applies, and the examples the policy will be tested against.

Keep the policy in draft until [simulation examples](/articles/simulate-a-policy-before-launch) match the owner’s intent. The plain-language boundary should remain understandable after the generated rule is reviewed.

## Related articles

- [Simulate a policy before launch](https://help.irongorilla.ai/articles/simulate-a-policy-before-launch)
- [Review an agent before deploy](https://help.irongorilla.ai/articles/review-an-agent-before-deploy)
- [Investigate blocked actions](https://help.irongorilla.ai/articles/investigate-blocked-actions)

## Support

Need help? Tell us where you got stuck: https://help.irongorilla.ai/contact