This repo packages two Agent Skills: Signal Geometry and Chalk Logic. Each skill gives your agent a focused visual style, example outputs, and prompt patterns for rendering or prompt-only use.
Builders who want their agent to produce specific illustration styles from a named skill.
You can ask for a consistent poster or chalk-style illustration without re-describing the visual rules each time.
Turns one concept into a sparse abstract illustration or poster using precise geometry, matte fields, restrained contrast, and one clear spatial event.
Turns one concept into a quiet white-chalk illustration on a desaturated green board, aimed at explanatory or editorial scenes.
The README shows using `$signal-geometry` and `$chalk-logic` directly in prompts.
Both skills support prompt-only use so you can generate the prompt and recipe without rendering an image.
Rendered runs return the accepted image plus the final prompt and composition recipe; Chalk Logic also returns QA status.
npx skills add CaliCastle/skills --skill signal-geometry
pnpm dlx skills add CaliCastle/skills --skill signal-geometry
pnpm dlx skills add CaliCastle/skills --skill chalk-logic
npx skills add CaliCastle/skills --skill chalk-logic
A collection of Agent Skills by Cali Castle. Currently includes Signal Geometry and Chalk Logic.
Signal Geometry is an Agent Skill for turning one concept into a sparse abstract illustration or poster. It uses precise geometry, quiet matte fields, restrained contrast, and one legible spatial event.
Using npm:
npx skills add CaliCastle/skills --skill signal-geometry
Using pnpm:
pnpm dlx skills add CaliCastle/skills --skill signal-geometry
Invoke the skill explicitly with $signal-geometry:
Use $signal-geometry to create a dark 4:5 poster about a weak signal becoming stable through repeated filtering. No text.
For a prompt without rendering:
Use $signal-geometry in prompt-only mode for an ultrawide illustration about two systems finding equilibrium.
Rendered work includes the accepted image, its exact final prompt, and the complete composition recipe. Prompt-only mode returns the prompt and recipe without generating an image.
$signal-geometry by name.Signal Geometry and its reference images are released under the MIT License.
Chalk Logic is an Agent Skill for turning one concept into a quiet, wordless white-chalk illustration on a desaturated green board. It supports explanatory systems, natural processes, observed objects, and sparse editorial scenes without drifting into classroom clutter or polished digital graphics.
Using pnpm:
pnpm dlx skills add CaliCastle/skills --skill chalk-logic
Using npm:
npx skills add CaliCastle/skills --skill chalk-logic
Invoke the skill with $chalk-logic:
Use $chalk-logic to create a quiet 16:9 wordless illustration about a heavy thought becoming easier to carry.
For a prompt without rendering:
Use $chalk-logic in prompt-only mode for a portrait illustration explaining how a key opens a lock. No labels.
Rendered work includes the inspected image, its exact prompt sequence, the complete composition recipe, and QA status. Prompt-only mode returns the compiled prompt and recipe without generating an image.
Chalk Logic and its reference images are released under the MIT License.
Sign in to join the discussion.
No comments yet. Be the first to say what this is good for.
Topic in, narrated explainer video out. A Claude Code / Codex skill that turns any topic into a black-canvas motion-graphics explainer video with TTS voiceover, subtitles and a chapter progress bar. Chinese or English; every frame drawn in code with Remotion.
Public repository for Agent Skills
An AI skill that provides design intelligence for building professional UI/UX across multiple platforms.
Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop
Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export.
Rebuild the object in a reference image as a code-only, procedural, quality-gated, animation-ready Three.js model. Token-efficient image-to-3D.