CaliCastle Skills
Two image skills for wordless visuals: sparse abstract geometry on matte paper and white chalk on a green board, one idea each.
- 98 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- The skill triggers image generation at the connected provider, and paid credit use depends on your plan
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add skills/calicastle-skillsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- The skill triggers image generation at the connected provider, and paid credit use depends on your plan
This is third-party code. Review the repository files before installing.
What it does
A repository of two skills that turn one concept into an illustration. Signal Geometry builds a sparse abstract composition: precise lines, a quiet matte field, one legible spatial event, in light or dark mode. Chalk Logic draws a wordless explanation in white chalk on a desaturated green board. When the agent has image generation and inspection, the skill renders the picture itself. Without them it returns the compiled prompt and composition recipe. Both modes include the exact final prompt and the full composition description.
Who it is for. Designers and writers who want restrained wordless illustrations and posters built around a single idea.
Good fit when
- A cover or slide about one concept, such as a weak signal stabilised by repeated filtering.
- A dark or light 4:5 poster with no text.
- A chalk explanation of a mechanism, such as how a key opens a lock.
- A prompt for an external generator: prompt-only mode returns it without rendering.
Not a fit when
- Labelled diagrams, charts with numbers and interface screenshots.
- Images with a logo, brand palette or recognisable character.
- A finished raster when the agent has no image tools: only the prompt will come back.
Example request
Use $signal-geometry to create a dark 4:5 poster about a weak signal becoming stable through repeated filtering. No text.Limitations
Rendering needs image generation and inspection in the agent, otherwise only prompt-only mode remains. Images carry no text by default: one short phrase or up to three labels totalling six words are allowed. In Codex, Signal Geometry is invoked by name only, as `$signal-geometry`. Each skill has a fixed style, so variety is limited to two manners. The result depends on the image model the agent uses, and on its cost. The last commit in the repository dates from July 2026.
How to disable. Delete the `signal-geometry` and `chalk-logic` folders from the agent skills directory. Images already created stay with you.
FAQ
How do the two skills differ?
Signal Geometry draws abstract geometric systems on matte paper in light and dark modes. Chalk Logic draws in white chalk on a green board and is closer to an explanatory sketch.
What does prompt-only mode return?
The compiled prompt and composition recipe without rendering. The mode works without image tools.
How do I invoke a skill in Codex?
Write its name with a dollar sign, for example `$chalk-logic`. The Signal Geometry Codex integration is explicit by design, with no auto-trigger.
Can labels be added to the image?
Yes, but sparingly: one short phrase or up to three labels, six words in total. There is no text by default.
What is the licence?
MIT, covering the skills and reference images.
Discussion
No questions or reviews yet.
Related
SkillsFive image-generation skills: Qwen-Image prompts, logos with shared geometry, diary comics, transparent PNGs and process dioramas
A skill that draws interactive isometric illustrations in SVG, with WebGL shaders and React output, explaining an idea as a working machine
Skill for editorial images in a minimal-zine style: paper, ink, series and a check of the finished bitmap.
A Codex skill that redraws your photo as a watercolor illustration on white paper, with wide margins and a handwritten caption
Skills give Claude Code three ways to make images: code-drawn graphics, Three.js 3D scenes and Flux diffusion on Cloudflare.
A Codex skill that turns a photograph into a vertical composition with an abstract memory panel and an English title
README in short
Each skill installs with its own command and the `--skill` flag. The README gives sample requests and describes what the output includes.