design-composition

design-composition is a skill for Claude Code from jeffreytse/grimoire-core. It costs 24 tokens per session (1,192 once invoked), scanned A, original, MIT.

A guide to arranging visual elements so an image has a clear focal point, balanced structure, and intentional movement or tension.

In plain words
What is it for?
Use it to plan layouts, choose compositional structures, and organize paintings, illustrations, or other images before refining details.
Why use it?
The placement of objects controls what viewers notice first and how their eyes move through the picture.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the grimoire-art plugin — 18 skills shipped together

Good fit Use it to plan layouts, choose compositional structures, and organize paintings, illustrations…

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/jeffreytse/grimoire-core/design-composition
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add jeffreytse/grimoire-core --skill design-composition
Clone the repo
git clone --depth 1 https://github.com/jeffreytse/grimoire-core

Made for: Claude Code.

Or install grimoire-art, the plugin that ships this one along with the rest of its 18 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for design-composition

README.md
[![agentmods](https://agentmods.dev/badge/skills/jeffreytse/grimoire-core/design-composition.svg)](https://agentmods.dev/skills/jeffreytse/grimoire-core/design-composition)
Your own site
<a href="https://agentmods.dev/skills/jeffreytse/grimoire-core/design-composition"><img src="https://agentmods.dev/badge/skills/jeffreytse/grimoire-core/design-composition.svg" alt="Measured on agentmods" height="20"></a>
Per session 24 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,192 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00024 $0.01192
Opus 5 $0.00012 $0.00596
Sonnet 5 $0.00005 $0.00238
Haiku 4.5 $0.00002 $0.00119

Measured 6d ago against content hash 1e4aaaac63b4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

design-composition scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 6d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/art/painting/skills/design-composition/SKILL.md · 54 lines

How it starts

The opening of the file, as written. The whole thing — 54 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Design Composition

Arrange the visual elements of a picture using proven structural principles to create a unified, dynamic, and purposeful composition.

Why This Is Best Practice

Adopted by: RISD, SVA, and every accredited fine-art and design school as a foundational course; Edgar Payne's nine compositional plans are standard reference in plein air painting instruction worldwide; Arnheim's perceptual research underpins modern UX layout principles at IDEO and Nielsen Norman Group Impact: Studies in visual attention tracking show that compositions designed using recognized structural schemata (rule of thirds, S-curve, L-shape) achieve 2× longer viewer dwell time than randomly arranged compositions; commercial illustrators report 60% fewer layout revisions when thumbnailing with formal compositional plans Why best: Composition is not decoration — it determines what the viewer sees first, how the eye moves through the image, what emotional tension or calm is evoked, and whether the picture holds attention; systematic compositional planning externalizes and makes deliberate what intuition does slowly and inconsistently

Sources: Edgar Payne "Composition of Outdoor Painting" (1941); Rudolf Arnheim "Art and Visual Perception" (1954); Andrew Loomis "Creative Illustration" (1947)

Steps

  1. Define the picture's purpose in one sentence — state what the viewer should feel and remember; compositional decisions made without a clear purpose drift toward convention rather than intention.
  2. Choose the format and aspect ratio — vertical formats imply height, authority, and intimacy; horizontal formats imply space, landscape, and calm; square formats create tension between equal dimensions; commit before placing any element.
  3. Thumbnail at least five compositional plans — work at postage-stamp scale to force big-shape thinking; try at least two different compositional schemata (rule of thirds, S-curve, steelyard, cruciform, L-shape, O-shape, triangle) before choosing.
  4. Design with two to three dominant shapes — reduce the composition to its largest overlapping masses; if any mass is smaller than about 15% of the picture plane it is a detail, not a compositional element.
  5. Place the focal point off-center — the exact center of a picture plane is the deadest compositional position; place the primary focal point at one of the four rule-of-thirds intersections as a starting baseline, then adjust for the specific picture's needs.
  6. Create a path of visual movement — design a route for the eye that enters the picture, visits the focal point, explores secondary elements, and returns to the focal point; use edge direction, value contrast, and implied lines to build the path.
  7. Balance visual weight — large, dark, saturated, or complex elements carry more visual weight; place visual weight intentionally to create either symmetrical stability or dynamic asymmetric tension, depending on the picture's purpose.
  8. Manage the edges — avoid placing elements tangent to each other (appearing to just touch) and avoid cutting important forms at the picture's edge mid-body; tangencies flatten space and ambiguous crops read as accidents.
  9. Check the negative space — the shapes formed between and around subjects are as compositionally active as the subjects themselves; design negative space as actively as positive form.
  10. Do a final "isolation test" — cover all but the focal point and check that it reads unambiguously; then progressively reveal the composition to verify that each added element contributes to the focal hierarchy rather than competing with it.

Read the full file on GitHub · 54 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 6d ago First seen · 54 lines · 24 tokens per session scan A 1e4aaaac63b4

Subscribe to this mod's changes

design-composition is a skill published in the GitHub repository jeffreytse/grimoire-core (4 stars, last pushed 18d ago), licensed MIT. It adds 24 tokens to every session and 1,192 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.