Hacker News
Show HN: Klad, a paint app your coding agent can drive (source-available)
I was looking for an application that would allow me to ask an agent to make a first draft of whatever art I needed with actual brushes, and then be able to go in myself and edit every last stroke or layer that it created.
Klad is that application. Written in Rust, it can be fully controlled by an agent through the CLI, and every function is available in a nice UI as well: a brush engine, frame-by-frame animation (and bones/IK), vectors, text layers, colours, layer effects, infinite undo (even after re-opening the file!),…
Why I think it's useful for a wider audience: Adobe showed agents editing Photoshop files at MAX this week, with a private beta planned for early 2027.
Klad is built on the same core idea, with editable paint strokes as core architecture, and it works today.
Clone the repo to try it out: https://github.com/klad-app/klad
You can basically point Claude to the repo, ask it to build it, and pick up the drawing skills. Then ask it to make any drawing you want! (After that, you or the agent can change absolutely every brush stroke in the app again, which makes edits a breeze.)
The quality of the drawings really depends on the agent you use. I have tried it with Claude Fable 5.2 since it came out on Monday, and the drawings are already quite impressive. I'm confident that this will only improve.
I use it every day on Mac, so that definitely works well. It builds on Linux too (tablet pen pressure still needs some work there). Windows builds in CI, I haven't used it there myself yet but I have some reports that it works perfectly fine.
Coming up next is support for pixel art. I'm in the testing phase with this right now, and I expect it to go live in the repo sometime this weekend.
Klad is source-available rather than open source: free to build and use for individuals and small companies, and I plan to sell pre-built binaries later, the way Aseprite does. The file format, CLI reference and agent skill are MIT.
What I want to hear from people who try it: Can your agent work out the CLI without help, and where does it get confused? Can you go in and change one of its strokes easily?
The command reference and guides are available at https://klad.studio
I'll be here answering any questions about the technical side as well, as it's a pretty interesting challenge to build a performant engine with editable brush strokes that render as paint.
vunderba
|next
[-]
So I've done a little bit of testing. When I loaded up the editor, I expected it to communicate with the installed binaries for my large language models, or possibly use a BYOK (bring your own key) approach. But once I understood that it's driven from the command line and you see the changes happen in the editor that gets spun up, I had no problems.
I've been experimenting a fair amount with LLMs driving procedural graphics myself, although my experiments have been mostly limited to relatively low-resolution stuff, such as EGA adventure game graphics.
So I decided to put Klad through a relatively hard test: I took a reference picture of an illustrated squirrel and asked it to recreate it, then set up the rigging for a basic walk cycle. This was using Fable 5.2 high.
Then I did the same thing, but with a more traditional approach, using standard generative models to manually create a walk cycle. This was all done relatively quickly, so bear that in mind.
Here are the results. I still think these are early days, and LLM-driven graphics are only going to improve.
And there are obvious advantages to a procedural system with rigging: Much easier to maintain consistency, palette control, and the ability to make precise alterations without having to deal with any kind of conventional in-painting nonsense.
ambewas
|root
|parent
[-]
I think the results are really interesting. I think it's very obvious the classic image-to-video approach yields much more natural results at this time and with the current models.
I think I'm going to do some testing of my own, and see where things can be improved. I do think that the bottleneck at this time, is the model you use itself.
But it's like you say, it's very early days with an LLM expressing its ability to draw/paint. In my own testing, it kind of feels like it did with coding 2 years ago: yeah sure the model is pretty Okay at building a function, but I wouldn't have let it build a whole feature back then.
Nowadays, I'm practically only reading code and thinking about how to refine my instructions for the next build.
I'm sort of betting that this evolution could go the same way for graphics (and possibly audio/midi as well). And I wanted to have a tool ready, to see how it evolves.
vunderba
|root
|parent
[-]
I definitely agree. For me, using an LLM to drive drawing tools for procedural graphics only really became viable in the last year with the advent of open-weight model Qwen3.8-27B and proprietary models like Opus 5x.
If you’re interested, I built something in a similar vein: a custom drawing language that renders old adventure game-style scenes and is also driven by an LLM.
It produces relatively decent results because the domain is inherently constrained to an EGA color palette and a low resolution designed to approximate old adventure games from the 1980s.
And like your tool - half the fun is watching the LLM drive the production of the final image in real time!
ImJasonH
|next
|previous
[-]
2001zhaozhao
|next
|previous
[-]
ambewas
|root
|parent
[-]
I explicitly designed the klad file format to be git compatible, which matters for agents that can easily track changes, but it is also a lightweight format that describes every brush stroke in a drawing rather than recording pixels.
The latter means every single one of the strokes an agent or you make, remains editable even thousands in. You can move every one, change the colour, transform it like a vector... And all the pixel rendering replays for the chunks it was in, in pretty much real time.
This then enables much deeper control for any changes you or an llm want to make, and I think this fundamentally makes klad positioned much better for agents drawing than any existing adobe clone.
I could be wrong about this, of course. Time will tell. For now I have received positive reports from people especially about the editability and consistency that comes with this format
BeetleB
|next
|previous
[-]
I like photography, but I'm relatively poor on the photo editing. I was hoping to build an MCP server around Gimp/Darktable/RawTherapee. The idea would be that the user would describe changes he wanted to see in the image (brighten up the shirt), and the LLM would pick appropriate tools to select the region and make the edit.
At the time I looked into it, it didn't seem that Darktable/RawTherapee provided APIs to hook into. Then life got busy.
ambewas
|root
|parent
[-]
I think this is much much more interesting compared to generative ai for images/photos/music/... I'm hoping more and more people will come up with alternatives to traditional software where an agent can actually really help you out, but where you stay in control
mmvvaa
|next
|previous
[-]
I envisioned it as some sort of a flipbook that would allow me to *cheaply* get a low-res idea of what the 'generate' click would get me, without having to burn movie making tokens.
This may not be exactly it, but it nudged me into a direction. Thanks!
I will give it a try and report back.
rchrp
|next
|previous
[-]
ambewas
|root
|parent
[-]
For example, but rough numbers here,.. the panel poster on the website took about 15 minutes of fable working it out, and the full scene with the creature following a traveler, was also about 10 minutes of ai work.
To me, it feels just about as fast as having a model code up a medium feature for an application. And that doesn't seem too strange, because I'm the end coding is kind of what it's doing with this app.
A benchmark I'd be really interested in, is how an agent would be able to clean up or improve a drawing made with pen tablet by a real artist. I've been focusing so much on the development side though that this is a test I have not been able to do, but I expect it to work really well. Every time I make an edit to a drawing myself, Claude just knows what I did and why.
chaostheory
|next
|previous
[-]
This is a really great tool that solves a specific pain point with image models though
ambewas
|root
|parent
[-]
I have no idea how an agent would leave a watermark in Klad. It's got no agent inside, and every mark that an agent made would have possibly been made by a human as well. I guess they would have to really specifically target Klad to be able to make a watermark happen...
arjgoel30
|next
|previous
[-]
studyturtlehq
|next
|previous
[-]
hvvivek
|next
|previous
[-]
ambewas
|root
|parent
[-]
I am still waiting on apple for signing a build, but once I can I will release binaries as well. There's an email list you can sign up on the site, if you're interested to be notified. Once the binary is out, an MCP will be as well. Right now it's a cli and a UI app you can build yourself.
But if you want to try a headless version, just point your favorite LLM to it and ask it to build it in a sandbox
theanonymousone
|next
|previous
[-]
ambewas
|root
|parent
[-]
The other thing is that this is also a first class drawing app with pen support. So you don't have to use an agent, and you can still profit from the editability. Or start a drawing yourself and have an agent come in later..
pierreb-aiva
|next
|previous
[-]
tamimio
|next
|previous
[-]
ambewas
|root
|parent
[-]
The underlying engine is not just vectors: a brush stroke is saved as data, and everything is rendered on the fly into pixels. This means that things like smudging and paint blending is possible, all the while being able to edit even the very first stroke you made.
The best mental model I have right now is: it's the editibility of vectors, with the flexibility of a digital paint brush
tescreal
|next
|previous
[-]
ambewas
|root
|parent
[-]
Should all be there in a license file on GitHub!