HearlBot

What it does

• Fetches the one address a listener added, following at most three redirects. It does not crawl, follow links or discover pages on its own.
• Reads text from HTML pages, PDFs, plain text and Markdown. Anything else is skipped.
• Uses the text only to write that listener's private episodes, which name your page as a source. Hearl does not publish, index or resell your pages.

How to recognize it

Every request, including the robots.txt check, sends this user agent:

HearlBot/1.0 (+https://hearl.app/bot)

Requests come from Google Cloud in the United States. We do not publish a fixed list of IP addresses.

The rules it follows

• HTTPS only. It never connects to private or internal network addresses, and it checks every redirect again.
• It reads your robots.txt before fetching a page and honours Disallow rules for HearlBot or, when there are none for it, for *. It matches paths by prefix and ignores Allow lines and wildcards, so a matching Disallow always wins. It keeps your robots.txt for up to an hour.
• It stops at 20 MB and after 15 seconds.
• It never logs in, never pays and never works around a paywall. Pages that ask for a login or a subscription are skipped.
• It does not fetch social networks, video and streaming sites, file lockers or known paywalled publications at all.

How to opt out

Add this to the robots.txt at the root of your site:

User-agent: HearlBot
Disallow: /

HearlBot picks the change up within an hour. You can also disallow only some paths. Listeners who try to add a blocked page are told the site asks not to be read automatically.

What HearlBot is not

Hearl's own research uses Google Search through Google's Gemini API. Those searches are made by Google's systems under Google's rules, not by HearlBot.

Contact

Questions or a problem with HearlBot? Email hello@hearl.app with the address of the page and the time of the request.