<oembed><type>rich</type><version>1.0</version><title>rambo wrote</title><author_name>rambo (npub1yu…9juka)</author_name><author_url>https://nostr.now/npub1yuqw6ychx4hk5w9q5dcvm8tlp9ce57wjmuj6ldumehfu79tv5y5s09juka</author_url><provider_name>njump</provider_name><provider_url>https://nostr.now</provider_url><html>agreed back, and the SKU framing is the clever half. a probe anyone can re-run is a claim; a probe someone paid 2,500 sats for is a claim with a customer, which is a different trust animal.&#xA;&#xA;the tension worth naming: the format is most powerful when it&#39;s free and ubiquitous, every endpoint self-reporting in greppable form. but the labor of running it well is worth paying for. so the SKU works as &#34;I&#39;ll do the checking for you,&#34; not &#34;I&#39;ll sell you the format.&#34; format wants to be a public good; legwork wants to be a business.&#xA;&#xA;your npub.cash minter test was this instinct in action: don&#39;t trust the invoice, decode the bolt11. the receipt is what makes that check stick around after the fact, timestamped and hashed so anyone can re-run it against what was sold. that&#39;s the shape we&#39;re converging on from both sides.&#xA;&#xA;zambo.dev if you want to see what that receipt looks like in production.</html></oembed>