agentsclimarketplace

Run2 anthropic brand tokens

Skill cxcscmu/SkillLearnBench/skills/b2-self-feedback-claude-sonnet-4-6/anthropic-poster-design/run2_anthropic-brand-tokens

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install
npx -y skills add cxcscmu/SkillLearnBench --skill run2_anthropic-brand-tokens

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

Precise Anthropic brand color tokens and typography guidelines for design artifacts, with matplotlib-specific notes.

SKILL.md

2.8 KB, 706 tokens by cl100k_base, as published. Nobody here has run it

Anthropic Brand Tokens (Refined)

Color Palette

Token NameDesign RoleHEX
Identity LightPage / poster background (never pure white)#FAF9F7
Corporate DarkOuter casing, headings, dark UI elements#1A1816
Primary Brand AccentConnectors, CTAs, interaction highlights#CF6E35
Secondary Brand AccentThermal management, supporting hardware#8C9FA0
Tertiary Brand AccentPCB substrate, circuit-board elements#7A8B72
Muted Mid GrayAnnotation leader lines, dividers#9B9B9B

Supporting Colors (derived, not brand tokens)

NameUsageHEX
Battery Warm NeutralBattery pack layer#C8BFB4
Metadata BarFooter/header subtle bar#EDEAE5

Color Rationale

  • Identity Light (#FAF9F7): warm off-white; the subtle sepia undertone prevents sterility
  • Corporate Dark (#1A1816): very dark warm charcoal; never cold pure black
  • Primary (#CF6E35): Anthropic's terracotta/coral signature — used sparingly for maximum impact
  • Secondary (#8C9FA0): muted teal-gray — a cool industrial complement for thermal hardware
  • Tertiary (#7A8B72): muted sage/olive — organic feel suits PCB green substrate
  • Mid Gray (#9B9B9B): exactly mid-brightness neutral for structural lines

Typography

matplotlib Font Guidance

  • System font for matplotlib on Linux: DejaVu Sans (always available, sans-serif)
  • Preferred when installed: Inter, Helvetica Neue, or Arial
  • Check available fonts: from matplotlib import font_manager; [f.name for f in font_manager.fontManager.ttflist]

Letter-spacing in matplotlib

  • IMPORTANT: letter_spacing is NOT a valid matplotlib Text property — it will raise AttributeError
  • Simulate letter-spacing by inserting spaces between characters: "N O V A" instead of "NOVA"
  • For large display titles, spaces of 1–2 characters work well

Text Hierarchy

LevelSizeWeightColor
Display title (NOVA)52–60ptboldIdentity Light on dark bg
Subtitle9–10ptlightMuted Mid Gray
Layer label9ptregularCorporate Dark
Annotation text8ptlightCorporate Dark
Metadata7.5ptregularMuted Mid Gray

Brand Constraints

  1. Background MUST be Identity Light (#FAF9F7), never pure #FFFFFF
  2. No neon colors, no high-saturation gradients
  3. Primary accent appears only for focal interaction points (not for backgrounds)
  4. Low saturation overall — muted, professional, warm-neutral aesthetic
  5. Dark bars (title/footer) use Corporate Dark, NOT pure black

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

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.