Seeing lots of posts about improved Claude Code responses by adding words like “ultrathink” to prompts There are even suggestions for this mentioned in Anthropic best practices docs Building off this idea, I’m excited to share my favorite new Claudius feature: Enhancements

Enhancements enable an easy & extendable UX to prompt injections, they can be set per session or globally for all sessions Screenshot below is the "Pirate Mode" Enhancement, just to show off an example of how they work 🏴‍☠️

Claudius comes with a set of 100+ preexisting Enhancements for both technical & non technical users Whether debugging a performance issue, analyzing a business plan, or making sure Claude Code tries its hardest when working on your problem - Enhancements can help with it all 🪄

In later builds, I also want to create a way for users to upload the Enhancements they find most helpful so other users can also take advantage & easily discover them 💡

And of course, auto optimizing prompts & picking the best model based on the content of the prompt are (hopefully) coming soon too...

Enhancements are just one example of some of the awesome things that can come out of this native application layer on top of Claude Code If you're interested in trying out Claudius, just reply here & I'll try to get you a build as soon as I can 🙏

Scan any documents, convert images into text, PDF files, etc. 👍

@AnthropicAI How do I get access?! Sign me up!

@AnthropicAI will text you!

@AnthropicAI Inte

@AnthropicAI huh?

@AnthropicAI Cant wait to try this!

@AnthropicAI will dm you!