> This page is for version πŸ‡¬πŸ‡§ UK. > For other versions, use one of these documentation indexes: > - πŸ‡¦πŸ‡Ί AU (default): https://docs.humanapproved.com/au/llms.txt > - πŸ‡¬πŸ‡§ UK: https://docs.humanapproved.com/uk/llms.txt > - πŸ‡ΊπŸ‡Έ US: https://docs.humanapproved.com/us/llms.txt > 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. # Agent detection > HumanApproved's agent detection at a high level, using your own bot signal, when to ask for approval, fail-open behaviour and privacy. Each check decides whether the purchase pauses. It looks for signs that an AI agent, rather than the customer, is completing the checkout. The decision is made by HumanApproved's servers, not in the page, and the response is only ever `continue` or `review`: the detail behind each check is in the portal. ## Two sources Each product uses one source, chosen on its **Agent detection** page. ### HumanApproved's checks The default. The embed looks at how the quote journey was used and at general device characteristics. HumanApproved adds what it can see about the request and its network, including agents that identify themselves. The checks are built into the embed. > **Tip** > > Put the embed on **every page** of the quote journey, in your shared layout, not just the payment step. The checks consider the whole journey from the first page the script is on. On the **Agent detection** page you can adjust the product's sensitivity and exclude groups of checks. Agents that identify themselves always count. ### Your own signal If you already run bot detection, such as your CDN's bot score or your own model, a product can use that instead. Pass the result as `signal`, to the embed or in the check's body: ```json { "quoteRef": "HTA-4471920", "signal": { "source": "cloudflare", "score": 2, "verifiedBot": "ChatGPT-User" } } ``` | Field | What it is | | ------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | | `source` | Where the signal came from, for example `cloudflare` or `mcp`, for your records. Letters, digits, `_`, `.` and `-`, up to 40 characters. If it's missing or doesn't match, it's recorded as `unknown`. | | `score` | Your detector's bot score, 0 to 100. | | `verifiedBot` | The verified bot's name, when your detector names one, for example `ChatGPT-User`. | | `isAgent` | Your own verdict. `true` for every purchase through an agent-only flow. | On the **Agent detection** page you choose how the product reads it: your verdict (`isAgent`), a verified bot, or a bot score at or below a limit you set. > **Warning** > > A signal passed through the browser can be altered by the agent driving it. Send it from your server where you can. ## When to ask for approval Whatever the source, the product's **When to ask for approval** setting, on its **Approval settings** page, decides what happens next: | Setting | What happens | | ---------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | | **Only when an agent is detected** | The checkout pauses when the check finds an agent. People buying for themselves go straight through. | | **Every purchase** | Every checkout pauses for the customer's approval, whoever is buying. For testing, demos, agent-only flows, or products where every sale needs it. | | **Never β€” monitor only** | Every checkout is checked and recorded, but none pauses: every check returns `continue`. | New products start as **monitor only**. Review **Agent detections** in the portal over a representative period, including runs with the agents you expect, then switch to **Only when an agent is detected** and publish. See [Monitor only](/uk/guides/monitor-only). ### From a server or agent flow With no browser to look at, HumanApproved's checks only see your server's own request. Decide on the product instead: set **Every purchase** if every purchase through the flow is an agent's, or use your own signal and send `"signal": { "source": "mcp", "isAgent": true }`. See [Server and agent flows](/uk/guides/server-and-agent-flows). ## Fail-open behaviour If a check gets no response within the configured timeout (default 2.5 seconds), or an error occurs, the embed lets the step proceed and reports the check as `fail_open`. The same applies if `quoteDetails` can't provide the quote. See [protect() options](/uk/embed/sdk/options) and [Handling the outcome](/uk/embed/sdk/on-result). ## Limitations Detection catches agents that identify themselves, automated browsers and most agents running in the cloud, but an agent designed to imitate a person may not be detected. Some real customers can look like agents too, for example someone using only a keyboard or a VPN. Keep your own controls at bind. ## Privacy * The checks measure behaviour and general device characteristics: counts and timings of actions on the page. They never read what's typed, which keys are pressed, where the pointer goes, or page addresses. * HumanApproved uses the request's IP address to look up the network it came from. * A check carries no personal data. The quote and mobile are sent only when the checkout pauses for an approval. * The embed keeps one session value and one random visitor ID in the browser. Name the checks in your privacy notice and cookie policy before you go live, as you would fraud prevention. See [Regions and data](/uk/concepts/regions-and-data).