Artifacts can now be 'slides' and 'design' types stored as project files, alongside the existing design-system type
What
- A new
slidesartifact type stores each slide as its own HTML file (project/slides/<id>.html), with aproject/deck.jsonindex describing the deck's order, sections and fonts. - A new
designartifact type stores each artboard as its own file underproject/, with aproject/canvas.jsonindex. - The instructions Claude Code gives the model for these file-based types were rewritten to explain that content goes into files under
project/, tracked with acreatedOnFilesmarker object, and published together in one step.
Why Previously, artifact types like these worked through an older "store document" model. Storing slides and designs as individual files with an index lets Claude Code build and update multi-file artifacts (a slide deck, a set of design artboards) more directly, and lets the model describe changes file-by-file instead of as one document blob.
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Claude applies a built-in design skill when it builds an artifact, so pages get a deliberate palette, typography, and layout without extra prompting. That skill also looks for an existing design system in your project before choosing its o…artifacts see the edit
While Claude works on your connector task, it is designed to take direction from you. Other people's messages in the thread reach Claude as information about the task, not as instructions. A message that tells Claude to change course or sh…claude-tag/concepts/personal-connectors see the edit
Anthropic's documentation has since written up design, on Personal connectors in channels.