Integrations

We will POST to your endpoint the moment an audit finishes, with the score, the counts and a link to the report.

Setup. Not switched on yet. When it ships you will add an endpoint URL in your workspace and receive the audit.completed event.

Start a scan on any URL and read the result, plus endpoints for agent-traffic, issues and friction data.

Setup. Authenticate with an API key in the X-API-Key header, or as a Bearer token. Keys are issued by us on request rather than self-serve.

Connect Agent Checker to Claude, ChatGPT or any MCP client and ask about your agent traffic in plain language. Six tools: list and create tracked sites, fetch an install snippet, query traffic, list issues, run a funnel.

Setup. Not available yet. When it ships you will point your MCP client at the endpoint and authenticate with the same API key.

A one-line tag for your site that shows which AI agents visit, what they do, and where they get stuck.

Setup. Not available yet. When it ships you will create a property and paste the tag before the closing body tag. No build step and no npm install.

Drift alerts when a previously passing check starts failing, plus your scheduled Quick check results and every finished report.

Setup. On by default for monitored sites. Pick the cadence per site in your dashboard.

Every audit exports as a PDF and a shareable client report page carrying your logo and colours, with no Agent Checker branding.

Setup. Upload a logo and pick your colours in settings. Applies to new reports immediately.