Skip to content

Introduction

Artbucket keeps a brand’s files and its rules in one place, and serves both to whoever asks: a person in the app, a script with an API key, or an agent in Claude, ChatGPT, Cursor or Figma.

Finding files is a solved problem. The hard part now is authority: which logo is current, which photo is still licensed for paid social in Germany, what the brand’s actual rules are. No agent can infer that from pixels. It has to be written down and served, and that is what Artbucket does.

  • One place for the current logo. Upload a new version once: embedded links, share links and the files on portals switch to it, and Artbucket shows you who still loads the old one so you can tell them.
  • A clear answer before a file goes out. Ask whether a photo may run in a country, a channel and on a date, and get a yes or no with the reason and a file to use instead.
  • Guidelines people actually open. Your rules become guideline pages on a portal, on your own domain if you like, released when you say so.
  • Files for agencies and partners without the back and forth. A link that ends on a date, a portal behind a password, or an upload link for work coming back to you for review.
  • AI tools that use the real brand. Claude, ChatGPT or Cursor read your rules and fetch the approved logo instead of guessing.
  • No lock-in. Titles, credits and tags are written into every file you download, and the brand can live as files in a Git repository too.

How it is built, for developers: How it works.

Artbucket is open source under the AGPL-3.0, copyright Pwnera SAS. Run it, change it, host it yourself, for any purpose; if you offer a changed version as a network service, publish your changes. The v1 API is frozen: see Stability.