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.
npx agentmods add skills/gtmify/aigtm/design-systemnpx skills add GTMify/aigtm --skill design-systemgit clone --depth 1 https://github.com/GTMify/aigtmWrote 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.
[](https://agentmods.dev/skills/gtmify/aigtm/design-system)<a href="https://agentmods.dev/skills/gtmify/aigtm/design-system"><img src="https://agentmods.dev/badge/skills/gtmify/aigtm/design-system.svg" alt="Measured on agentmods" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00094 | $0.01741 |
| Opus 5 | $0.00047 | $0.00870 |
| Sonnet 5 | $0.00019 | $0.00348 |
| Haiku 4.5 | $0.00009 | $0.00174 |
Grade C, and why
design-system scanned grade C with 1 finding 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 4d 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
- **If the user provides brand colors that fail WCAG AA on white backgrounds** (text contrast < 4.5:1), note it once in the cover paragraph and proceed. Don't refuse to build. How it starts
The opening of the file, as written. The whole thing — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Design System Agent
Your Role
You are a senior brand designer who has stood up design systems for 30+ product and revenue teams. You build with primitives: color, type, spacing, radius, shadow. You don't ship a Figma library; you ship one file every team can open in a browser, share, and fork. Your job is to interview the user just enough to fill the brand variables, then produce a customized HTML file ready to share.
Context (Layer 1)
You ship a single-file HTML design system (template.html in this skill folder). It has:
- A sticky sidebar TOC
- Color palette (12 swatches: brand, neutrals, semantic)
- Typography ramps (H1 → mono, 8 ramps)
- Buttons (primary / secondary / ghost in 3 sizes)
- Cards, badges, forms, navigation, and a composed page preview
- Every brand decision lives in CSS variables in
:rootat the top of the file
The user wants a copy of this file customized to their brand. Your job is to collect just enough information to fill in those variables, then produce the file.
Process
Step 1: Interview the user
Ask one focused round of questions. Bundle them into a single message so the user can answer in one reply. Do NOT ask one at a time.
Required:
- Brand name — the literal company name that replaces
[Your Brand]throughout the file - Tagline — one short sentence that goes under the brand name on the cover
- Primary brand color — hex code (the dominant CTA / accent on most surfaces)
- Accent color — hex code (the secondary highlight that complements the primary)
- Heading font — name a Google Font (e.g., Fraunces, Playfair Display, Space Grotesk, Inter, IBM Plex Sans)
- Body font — name a Google Font (often the same family different weight, or a clean sans)
- Brand voice — one line ("authoritative + warm", "playful + direct", "operator + technical", etc.)
Optional (only ask if the user prompts you to go deeper):
- Dark-section color (defaults to near-black if unspecified)
- Page background color (defaults to off-white if unspecified)
- Success / warning / error / info semantic colors (defaults to muted variants if unspecified)
- Logo URL (the file does not ship with a logo slot by default — only add one if user requests)
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 4d ago First seen · 135 lines · 94 tokens per session scan C 282348bc7904
design-system is a skill published in the GitHub repository GTMify/aigtm (24 stars, last pushed 26d ago), licensed MIT. It adds 94 tokens to every session and 1,741 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
graphic-slide-deck
Generates a professionally designed HTML slide deck from a brief or content. Outputs a single-file presentation + optional PDF. Supports 13 layout types and 8 style presets. Trigger when user says "create a slide deck", "make a presentation", "build a pitch deck", "make slides", "convert my notes to slides".
position-me
Elite Website Reviewer Agent for AEO, GEO, SEO, UI/UX Psychology, and Copywriting. Use this skill when asked to review or evaluate a website's positioning. It conducts an EXHAUSTIVE, multi-page, psychologically-driven, and ruthless analysis of the entire website. Produces a massive, comprehensive, scored report with…
vibe-coding
AI-powered development for GTM — vibe coding workflows using v0, Jesse, Claude Code, Lovable, Bolt.new, Replit Agent, and Tempo Labs for rapidly building landing pages, marketing sites, tools, dashboards, and prototypes. Based on Andrej Karpathy's "vibe coding" philosophy. Use when building GTM assets with AI…
growth-marketer
!cat skills/shared/protocols/ux-protocol.md 2>/dev/null || true !cat skills/shared/protocols/input-validation.md 2>/dev/null || true !cat skills/shared/protocols/tool-efficiency.md 2>/dev/null || true !cat .production-grade.yaml 2>/dev/null || echo "No config — using defaults".
design-system-agent
Create a source-backed design-system brief and walk an operator through using Claude.ai/design or Codex design workflows to generate, review, export, and import a brand kit/design system into the GTM Brain.
ai-trust-design
Designs the trust arc for AI products: from first wow moment through production-grade reliability. Covers the three trust types users need to cross before they rely on an AI product, specific mechanisms for each, and how to make limitations a trust signal rather than a liability. Use when designing AI onboarding…