---
title: "Discussion on \"Anthropic Ran a Real Agent Economy Inside Their Company. Here's What It Proved About Communication.\" | Hashnode"
description: "Discussion on \"Anthropic Ran a Real Agent Economy Inside Their Company. Here's What It Proved About Communication.\". Anthropic's Project Deal showed 69 employees' agents completing 186 deals over $4,000. What it actually proved: Slack doesn't scale to agent economies."
url: "https://hashnode.com/posts/anthropic-ran-a-real-agent-economy-inside-their-company-heres-what-it-proved-about-communication/69ee1ac0817a4c0e12e08ecf"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [Anthropic Ran a Real Agent Economy Inside Their Company. Here's What It Proved About Communication.](https://kavinkimcreator.hashnode.dev/anthropic-ran-a-real-agent-economy-inside-their-company-heres-what-it-proved-about-communication?utm_source=hashnode&utm_medium=feed)

Anthropic's Project Deal showed 69 employees' agents completing 186 deals over $4,000. What it actually proved: Slack doesn't scale to agent economies.

Apr 26 · 5 min read · kavinkimcreator.hashnode.dev  
Read the article

-   [#agents](https://hashnode.com/tag/agents)
-   [#ai](https://hashnode.com/tag/ai)
-   [#programming](https://hashnode.com/tag/programming)

## Written by

[kavin-kim-creator](https://hashnode.com/@kavin-kim)

@kavin-kim

  
[View profile](https://hashnode.com/@kavin-kim)

## Responses 1

Markdown supported · ⌘↵ to post

[Ethan Frost](https://hashnode.com/@ethanfrost)

[Apr 27](https://hashnode.com/posts/anthropic-ran-a-real-agent-economy-inside-their-company-heres-what-it-proved-about-communication/69ee1ac0817a4c0e12e08ecf/comment/69ef1cb2dc472baafda348f1)

the part that surprised me from this write-up: a lot of the friction was about agents not knowing what _not_ to do, even more than knowing what to do.

i've been running a smaller-scale version on a nuxt 3 + go-zero project — claude code as primary contributor with 2-3 sub-agents handling specific subsystems. the thing that scaled this for me wasn't better prompts. it was treating AGENTS.md / CLAUDE.md / skill files as the actual coordination protocol. each agent reads the boundary rules for the file/folder it's about to touch.

basically: a lightweight "right of way" system. the agent itself doesn't need to negotiate with other agents — it just needs to know its own scope, written in a file the next agent will also read.

been keeping mine + others' AGENTS.md / skill files at [https://tokrepo.com/en](https://tokrepo.com/en). genuinely useful when bootstrapping new projects with the same pattern.
