Borrowing it
Nothing to install: this file belongs to YuqingNicole/variant-design-skill. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/YuqingNicole/variant-design-skill/master/AGENTS.mdgit clone --depth 1 https://github.com/YuqingNicole/variant-design-skillWrote 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/instructions/yuqingnicole/variant-design-skill/agents-md)<a href="https://agentmods.dev/instructions/yuqingnicole/variant-design-skill/agents-md"><img src="https://agentmods.dev/badge/instructions/yuqingnicole/variant-design-skill/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/instructions/yuqingnicole/variant-design-skill/agents-md"><img src="https://agentmods.dev/badge/instructions/yuqingnicole/variant-design-skill/agents-md.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.02377 | $0.02377 |
| Opus 5.5 | $0.00951 | $0.00951 |
| Sonnet 5.5 | $0.00475 | $0.00475 |
| Haiku 4.5 | $0.00238 | $0.00238 |
Grade A, and why
variant-design-skill AGENTS.md 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 today.
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.
How it starts
The opening of the file, as written. The whole thing — 101 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Guidance for coding agents (Codex, Claude Code, others) working on this repository. To use the skill for design work in another project, read SKILL.md instead.
What this repository is
Variant Design is an agent skill. It turns one brief into three comparable design directions (A/B/C) inside the user's own project. It supports scoped edits with snapshots and undo, and integrates the chosen direction with a rollback path. There is no hosted service: the user's agent supplies the model, and the bundled Node scripts do the stateful work.
This directory is itself an installed skill. Users clone it whole into ~/.codex/skills/variant-design or ~/.claude/skills/variant-design, and the maintainer's own install may be a symlink to this working tree. Every file you change here takes effect in live agent sessions immediately.
Layout
| Path | Role | Notes |
|---|---|---|
SKILL.md |
Router entry point | Keep it short. It routes to one sub-skill; do not inline workflows here |
skills/*/SKILL.md |
Sub-skills: generate, design-system, component, analyze, ux | Shared phases each have one source of truth: skills/shared/code-output.md (writing code), references/workflow/preview.md, references/workflow/history.md and skills/shared/export.md. Each route starts with a load map; files are read at the step that needs them. Only the root SKILL.md has skill frontmatter: Codex and npx skills scan folders recursively and would register sub-skills as broken standalone skills (enforced by plugin-package.test.py) |
references/ |
Loaded on demand by the routes | Grouped by role: workflow/ (context, discovery, history, quality and integrity rules shared by routes), sources/ (design references and direction cards), domains/, design-system/, ux/, tasks/ (routes the router sends straight to a reference), plus the shared palettes.md and interactive-patterns.md. Domain references follow the five-part structure in README.md → Contributing. scripts/skill-links.test.mjs fails on a named file that does not ship with the skill and on a reference no route reaches |
scripts/*.mjs |
Runtime helpers that run inside user projects | Zero dependencies, Node 22+, ESM. Each has a CLI and exported functions |
scripts/*.test.mjs |
node:test unit tests for the helpers |
Excluded from the plugin package |
tests/ |
Playwright browser regressions (loop.spec.mjs, landing.spec.mjs) |
Has its own package.json |
website/ |
Bilingual React + Vite marketing site and online lab | Also the dogfood project: website/variant-output/ holds real A/B/C history. website/content/ holds direction-library entries (RFC-002), validated by website/scripts/content.mjs |
mcp/ |
Local stdio MCP adapter over the same history helpers | Has its own package.json and tests |
evals/ |
Real agent runs on fixture projects, graded automatically | See evals/README.md. Runs cost model usage; the grader tests do not |
plugin/ |
Codex plugin manifest and release-readiness notes | Built by scripts/build-plugin.py |
docs/prd, docs/rfc |
Requirements and architecture, written in Chinese | GitHub Issues track status. Writing a PRD is not shipping it |
docs/research/ |
Dated dogfood records and competitor notes | Record evidence here; do not rewrite history |
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.
- today Changed · +5 lines · +487 tokens per session 7e22d15186ec
- today First seen · 96 lines · 1,890 tokens per session scan A 1b30b0bc5f72
variant-design-skill AGENTS.md is an instructions file published in the GitHub repository YuqingNicole/variant-design-skill (47 stars, last pushed today), licensed MIT. It adds 2,377 tokens to every session, about $0.0095 per session on Opus 5.5. 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 2026-10-27.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
deno copilot-instructions.md
Copilot instructions for denoland/deno, covering deno development guide for github copilot, network access, high level overview, key directories and key files to understand first.