> This is HumanApproved's documentation, as Markdown for AI agents. /llms.txt lists every page; /llms-full.txt has them all. If you're helping someone integrate HumanApproved, start with /guides/ai-coding-agents.

# Concepts

> Pause an AI agent's insurance purchase until the customer approves it.

## Docs

- [Checks and reviews](https://docs.humanapproved.com/us/concepts/checks-and-reviews.md): How HumanApproved checks, outcomes, review statuses, expiry, decline reasons, results and product versions fit together.
- [Approval methods](https://docs.humanapproved.com/us/concepts/approval-methods.md): The HumanApproved approval page, the SMS one-time code method and its limits, the optional identity check, and what an approval records.
- [Agent detection](https://docs.humanapproved.com/us/concepts/agent-detection.md): HumanApproved's agent detection at a high level, using your own bot signal, when to ask for approval, fail-open behaviour and privacy.
- [Keys and environments](https://docs.humanapproved.com/us/concepts/keys-and-environments.md): HumanApproved publishable keys, review tokens and secret keys, test and live environments, allowed websites and the Origin header, and the host each key works with.
- [Regions and data](https://docs.humanapproved.com/us/concepts/regions-and-data.md): HumanApproved regions and their hosts, the data collected by checks and reviews, how the mobile is protected, and how to keep userId opaque.