I can't even tell if this repository is based on prompts extracted from Claude Design or if the author had an LLM create all of these prompts in it from scratch.
The fact that they encourage and accept PRs indicates that this isn't intended as a direct prompt extraction exposure project - plus the license, which should indicate they have the authorship necessary to license that content.
Assuming this IS a complete ground-up implementation it really needs to link to demonstrations that it works. Without any evidence it's hard to justify spending time exploring it.
If you ask Claude Design itself to list the names of the skills available to it you get:
Animated video
Interactive prototype
Make a deck
Make a doc
Make tweakable
Claude API in prototypes
Frontend design
Wireframe
Export as PPTX (editable)
Export as PPTX (screenshots)
Create design system
Save as PDF
Save as standalone HTML
Send to Canva
Handoff to Claude Code
Which does not match the structure of this project at all.
This would be much more interesting if it detailed how the prompt/skills were reverse-engineered. As it is it seems like this could just be the output from “hey Claude write me a system prompt that works like Claude Design”.
They have never attempted to hide their system prompts in fact they explicitly publish them for Claude (https://platform.claude.com/docs/en/release-notes/system-pro...) and Claude Code they provide a proxy option which then makes it trivial to see the full requests and responses including the prompts and tool usage...
honestly, i think you can just look at the network tab and see the "content" of the skills. Same has been true for their excel addin and bunch of other things.
I've been using Claude Design to make animated SVGs, and I've learned a thing or two about its limits and how to get around them for that narrow purpose.
One thing I've learned is that you have to ask it to first come up with a robust way to define the geometry and then apply that to an SVG. Without that first step, it just guesses at where everything should be that isn't directly connected with a node, and it is hilariously bad. But with that first step it is capable of creating some incredible geometry algorithmically from detailed instructions.
The other thing is that whatever tool prepares the svg for export will strip the animations as part of a sanitizing process, it won't even see that has occurred. You have to ask it to export to a different file type like my-animation.svg.txt, and then obviously you want to inspect it because svg can carry many exploits not related to animation.
Its ability to generate these designs is part toddler with a crayon, part savant. It won't be able to create any quality organic figures without reference material, but its ability to create intricate and mathematically correct design and animate it is incredible. I haven't used it extensively so I'd be very interested to hear other observations and advice.
If this is regular output of the LLM, I'm not sure, but given that the author proclaims that this is reverse engineered, then they are not allowed to redistribute it under their own license terms. The terms of service are also pretty clear on this not being allowed, which makes it extra hard to defend (section 3.3):
> You may not access or use, or help another person to access or use, our Services in the following ways:
...
3. To decompile, reverse engineer, disassemble, or otherwise reduce our Services to human-readable form, except when these restrictions are prohibited by applicable law. [1]
How can you tell what regular output is? Is there a special output when you successfully jailbreak? Is there a meaningful distinction between jailbroken prompts and hallucinations? Are certain prompts against the terms of service? If so, is it easy to determine if they are? Who determines this? If you produce output that is against the terms of service, does that change the copyright status of the works?
I'm calling BS, sorry. It looks light, and barely anything beyond surface level of what we could all could guess would be in a system prompt. This smells nothing more of a "claude give a system prompt that anthropic would use as a system prompt for claude"
From what we know, there are some very specific details baked into the prompt as safety guards, where are those? Again calling BS and I'm not gonna waste more thought/words on this
I can't even tell if this repository is based on prompts extracted from Claude Design or if the author had an LLM create all of these prompts in it from scratch.
The fact that they encourage and accept PRs indicates that this isn't intended as a direct prompt extraction exposure project - plus the license, which should indicate they have the authorship necessary to license that content.
Assuming this IS a complete ground-up implementation it really needs to link to demonstrations that it works. Without any evidence it's hard to justify spending time exploring it.
If you ask Claude Design itself to list the names of the skills available to it you get:
Which does not match the structure of this project at all.There's this one which seems like it matched it: https://github.com/jimliu/baoyu-design
Yeah these skills match that list of names, looks like this one may have extracted all of the prompts: https://github.com/JimLiu/baoyu-design/tree/main/skills/baoy...
Agree, it’s vibe slopped rather than the actual Claude design system prompt
Claude design's prompt is trivial to verify. They bundle it in the frontend bundle and send it on every network request.
This would be much more interesting if it detailed how the prompt/skills were reverse-engineered. As it is it seems like this could just be the output from “hey Claude write me a system prompt that works like Claude Design”.
> This would be much more interesting if it detailed how the prompt/skills were reverse-engineered
That would enable Anthropic to block the technique.
They have never attempted to hide their system prompts in fact they explicitly publish them for Claude (https://platform.claude.com/docs/en/release-notes/system-pro...) and Claude Code they provide a proxy option which then makes it trivial to see the full requests and responses including the prompts and tool usage...
honestly, i think you can just look at the network tab and see the "content" of the skills. Same has been true for their excel addin and bunch of other things.
I've been using Claude Design to make animated SVGs, and I've learned a thing or two about its limits and how to get around them for that narrow purpose.
One thing I've learned is that you have to ask it to first come up with a robust way to define the geometry and then apply that to an SVG. Without that first step, it just guesses at where everything should be that isn't directly connected with a node, and it is hilariously bad. But with that first step it is capable of creating some incredible geometry algorithmically from detailed instructions.
The other thing is that whatever tool prepares the svg for export will strip the animations as part of a sanitizing process, it won't even see that has occurred. You have to ask it to export to a different file type like my-animation.svg.txt, and then obviously you want to inspect it because svg can carry many exploits not related to animation.
Its ability to generate these designs is part toddler with a crayon, part savant. It won't be able to create any quality organic figures without reference material, but its ability to create intricate and mathematically correct design and animate it is incredible. I haven't used it extensively so I'd be very interested to hear other observations and advice.
> Open source, MIT licensed.
I don't think that is how copyright licensing works.
Isn't it, though? What's the copyright status of the output of these tools?
If this is regular output of the LLM, I'm not sure, but given that the author proclaims that this is reverse engineered, then they are not allowed to redistribute it under their own license terms. The terms of service are also pretty clear on this not being allowed, which makes it extra hard to defend (section 3.3):
> You may not access or use, or help another person to access or use, our Services in the following ways:
...
3. To decompile, reverse engineer, disassemble, or otherwise reduce our Services to human-readable form, except when these restrictions are prohibited by applicable law. [1]
[1] https://www.anthropic.com/legal/consumer-terms
How can you tell what regular output is? Is there a special output when you successfully jailbreak? Is there a meaningful distinction between jailbroken prompts and hallucinations? Are certain prompts against the terms of service? If so, is it easy to determine if they are? Who determines this? If you produce output that is against the terms of service, does that change the copyright status of the works?
I'd love to see this go to court.
The terms of service are between Anthropic and one of their subscribers. So Anthropic can maybe cancel their contract.
This doesn't affect what copyright law allows or does not allow.
Also I think Anthropic very much suppports gathering data by whatever means possible. That should work both ways.
How to prove this is indeed the system prompt against a certain timestamp (in the past ) ?
I trust on this: https://github.com/elder-plinius/CL4R1T4S/blob/main/ANTHROPI...
It’s different than this one shared by the op, but Anthropic maybe updated the prompt
This is pretty awesome. I’ve wanted to use Claude design, but with my regular MCP servers.
Side note: ironic use of an llm writing the readme.
I'm calling BS, sorry. It looks light, and barely anything beyond surface level of what we could all could guess would be in a system prompt. This smells nothing more of a "claude give a system prompt that anthropic would use as a system prompt for claude"
From what we know, there are some very specific details baked into the prompt as safety guards, where are those? Again calling BS and I'm not gonna waste more thought/words on this