Why We Built ACP and Why We're Open Sourcing It

Episode stack: https://stackl.ist/4thLgx3 The internet was built for humans. But the fastest-growing users of the internet are now AI agents. They read the same pages over and over, burning tokens to figure out what each piece of content is. There's no shared language for any of it. In this episode of Building Stacklist, we explain ACP (Atomic Content Protocol), the open standard we built. ACP is why a Stacklist card is already structured the way agents want to read it. The conversation walks through why a protocol matters more than another app. What 84% token savings looks like in a live benchmark. And why standardization at the content layer is the kind of infrastructure that has to exist before agentic search and discovery can truly work at scale. Topics - Why the internet's current structure breaks down once agents are doing the reading - The SMTP analogy: how a shared protocol turned email from walled gardens into a network - Why ACP is a protocol, not an app - The 84% token savings benchmark, demoed live - JSON-LD is for crawlers; ACP is for agents (and why both layers matter) - Building Stacklist on ACP: cards, stacks, and search - How to start using ACP today Chapters - [00:00] Intro - [01:25] The dumpster fire that is the downloads folder - [07:40] Why a protocol, not an app (SMTP analogy) - [09:19] Internet was built for humans; agents are the new users - [10:51] The 84% token savings benchmark, demoed live - [15:31] What this means for Stacklist users - [19:20] Inside an ACP-enriched card and stack - [23:04] What the world looks like if ACP wins - [24:22] Where to start Links to use ACP - Atomic Content Protocol: https://atomiccontentprotocol.org - NPM package: https://www.npmjs.com/settings/atomic-content-protocol/packages