agentsclimarketplace

Academic figure color expert

Skill Azhi-ss/academic-figure-skills/academic-figure-color-expert

AI skills for academic paper figures: repo analysis, figure planning, colorblind palettes, JSON figure specs. Install: npx skills add Azhi-ss/academic-figure-skills

Install
npx -y skills add Azhi-ss/academic-figure-skills --skill academic-figure-color-expert

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Palette Decision for academic figures — venue-aware style family and colorblind-safe scheme selection with scene recipes. Use when the user needs an academic color palette, classic vs pastel advice, Nature Blue, or architecture-diagram color matching.

SKILL.md

3.1 KB, as published. Nobody here has run it

Academic Figure Color Expert

Produce a reusable Palette Decision. Hex tables, venue maps, and scene→style→palette recipes live in one place:

→ Read ../docs/palettes.md (presets + Style family + Scene → palette decision + recipes).

Always decide style family (classic vs pastel) before naming a classic preset.

Principles

  1. Color carries information, not decoration.
  2. ≤ 3 chromatic colors + neutrals per figure.
  3. Colorblind-safe by default; dual-encode categories.
  4. ≥ 4 modules → monochrome (Nature Blue) beats polychrome busyness.

Input Contract

  • Prefer: venue, domain, figure type, module count, reference image, accessibility / print constraints
  • Minimum: any one of venue / figure type / domain
  • Missing info: follow ../docs/missing-info-policy.md; still emit a conservative Palette Decision

Output Contract — Palette Decision

Always include:

  • recommended palette + one alternate
  • primary / secondary / tertiary (+ neutrals) hex
  • reason (venue / domain / module count)
  • accessibility note
  • handoff block ready for prompt skills

Steps

Step 1: Collect constraints

Done when you have recorded (or marked missing): venue, domain, figure type, module count, colorblind/print needs, user color preference, reference image cues.

Step 2: Decide style family + palette

  1. Classic vs pastel — docs/palettes.md Style family first
  2. If pastel → hand off scheme P1/P2/P3 to academic-figure-prompt-pastel (hex from that skill)
  3. If classic → apply Scene → palette decision (hard constraints → figure type → venue → domain → vibe)
  4. Name primary + alternate; state branch (user / scene / default)

Done when: family + primary + alternate are explicit, with the decision checklist fields from docs/palettes.md.

Step 3: Emit hex + handoff

Load the chosen preset from docs/palettes.md. Output the Palette Decision format.

Done when: every role hex is filled, accessibility stated, and the handoff block is copy-ready for academic-figure-prompt.

Sparse-input cases

caseaction
no venueinfer from domain / reference; else Okabe-Ito (or Nature Blue if ≥ 4 modules)
accessibility unspecifiedassume colorblind-safe required
only vibe words (“高级/科技/柔和”)map to 1–2 presets with concrete hex
only reference imageextract hues; academicize if needed (white fill, border color, ≤ 3 chromatics)
figure type unknowndefault framework advice; note module-detail vs comparison alternates

Stop

Stop when the user has a Palette Decision for the current figure, or when zero constraints and zero artifacts exist (then ask for venue / figure type / domain).

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.