Skip to Content
For CreatorsAI agents

AI agents

Shosho runs on open protocols and public APIs, so you don’t have to do everything by hand. If you use an AI agent — the kind of assistant that can run commands and browse the web — you can ask it to do things on Shosho for you.

It all hangs on one link: shosho.live/llms.txt . That’s a plain-text guide written for AI agents. Tell your agent to read it and follow it, and it knows how to work with Shosho on your behalf.

What an agent can do for you

  • Get you live — set up your streaming details and go live, including from OBS or your own encoder.
  • Set up your profile — create or update your profile, picture, and relays.
  • List products — add Classified products (buyers press View on Website) or Shopping Cart products (buyers check out on Shosho) in your shop.
  • Set up a card break — one product with a spot per team, priced and stocked, ready to sell.
  • Post clips — publish a video clip to your profile.
  • Join the conversation — post chat messages and reactions on a live stream.

How to use it

Pick an agent

Use an AI agent that can run shell commands and access the web — many coding assistants can.

Point it at the guide

Tell it to read shosho.live/llms.txt and follow it for your task. For example:

Review the result

Everything the agent does appears under your profile on Shosho — check it in the app, on the web, or in Shosho Seller Hub.

A few things to know

  • New to Shosho? An agent can set up a Shosho identity for you as part of the task. Your identity is a Nostr keypair — see Public and private keys to understand and protect it.
  • Opening a Shopping Cart shop still needs us. An agent can build your shop and its products right away, but a Shopping Cart shop only becomes buyable once the Shosho team moves it to a published state.
  • You stay in control. Everything the agent does lives on the open network under your profile — you can review, edit, or remove it any time in the Shosho app, on the web, or in Shosho Seller Hub.

For developers

Want the detail behind llms.txt? See the developer docs, including example recipes for agents.

Last updated on