Everything an audit produces is available over an API, so results can land in whatever your team already uses. Keys are issued on request rather than self-serve; ask support and we will set one up.
Authenticating
Every request carries your key. Both header forms are accepted:
curl https://go.agentchecker.ai/api/v1/scan \
-X POST \
-H "X-API-Key: ak_live_your_key_here" \
-H "Content-Type: application/json" \
-d '{"url":"https://example.com"}'Or as a bearer token, if that fits your client better:
-H "Authorization: Bearer ak_live_your_key_here"The full key is shown once, at creation. Store it somewhere you can retrieve it, because we cannot show it again.
Rate limits
Each key is limited per minute. Every response carries the limit and remaining count in headers, and exceeding it returns a 429 with a Retry-After telling you how long to wait. Read the limit from the headers rather than hard-coding it.
Getting results back
Rather than polling, register a webhook. We POST to your endpoint when an audit finishes, with the score, the counts and a link to the report. The event is audit.completed.
That webhook is also how you reach anything we do not integrate with natively, including Slack and Zapier. See the integrations page.
The MCP server
If you would rather ask questions than write requests, the same key authenticates an MCP endpoint. Point Claude, ChatGPT or any MCP client at it and you can create a tracked site, fetch its install snippet, and query agent traffic, issues and funnels in plain language.