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 skills add yike-gunshi/forge-skills --skill forge-designgit clone --depth 1 https://github.com/yike-gunshi/forge-skillsWrote 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/yike-gunshi/forge-skills/forge-design)<a href="https://agentmods.dev/skills/yike-gunshi/forge-skills/forge-design"><img src="https://agentmods.dev/badge/skills/yike-gunshi/forge-skills/forge-design/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/skills/yike-gunshi/forge-skills/forge-design"><img src="https://agentmods.dev/badge/skills/yike-gunshi/forge-skills/forge-design.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 | $0.00120 | $0.04078 |
| Opus 5.5 | $0.00048 | $0.01631 |
| Sonnet 5.5 | $0.00024 | $0.00816 |
| Haiku 4.5 | $0.00012 | $0.00408 |
Grade A, and why
forge-design 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 19d 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.
How it starts
The opening of the file, as written. The whole thing — 320 lines — stays where its author put it; the contents beside it link to each section on GitHub.
文档落地路径：遵循 forge-doc-policy 规范。完整白名单 + frontmatter schema 见
~/.claude/skills/forge-doc-policy/doc-paths.md。 当前文档加载契约：先读项目CLAUDE.md、docs/README.md、docs/INDEX.md和docs/DESIGN.md当前真相源；长 changelog、Image2 历史稿和 raw archive 只在追溯原因时加载。详见~/.claude/skills/_shared/current-doc-loading.md。
/forge-design：全栈设计规划与文档管理
纯设计规划——不写代码。产出 DESIGN.md 和设计方案，交由 forge-design-impl 或 forge-eng 实现。
流程总览
门控判定（L1/L2/L3）
│
├── L1 完整设计 ──→ 竞品调研 → 美学方向(3+) → 配色/字体 → Token体系 → forge-design-lab 工作台（布局/比例/密度）→ 完整审计 → 0-10审查 → DESIGN.md
├── L2 轻量审查 ──→ 读DESIGN.md → 对照检查 → 重点审计(10项) → 涉及布局/组件时调 forge-design-lab → 一致性验证 → 更新文档
└── L3 跳过设计 ──→ 纯后端/无UI → 直接进 forge-eng
全程中文。关键设计决策需用户确认后才能定稿。
涉及前端页面、组件、状态或布局时，读取 ~/.claude/skills/_shared/visual-decision-layer.md：布局、比例、密度、栏宽、一屏几张这类决策走 forge-design-lab 工作台门禁（真 token、真像素的骨架，用户在上面改值、判断、保存）；Image 2 只用于气质与风格方向的预判，不用它判断比例。
分级门控体系
L1 完整设计（不可跳过）
触发条件：
- 新项目、新页面、新的独立功能模块
- 迭代不满触发器命中（历史账本显示同一模块 ≥ 3 次反复迭代）
流程： 第0步 → 第1步（创建模式）→ 第2步 → 第3步 → 第4步 → 第5步 → 第6步 硬门控： 第5步评分 < B 必须返工。子模块必须通过与父页面的一致性检查。
L2 轻量审查
触发条件：
- 迭代已有功能、样式微调、小组件添加
流程： 第0步 → 第1步（迭代模式）→ 第2步（重点10项）→ 第3步（简化）→ 第4步 → 第5步 门控： 评分 < C 升级到 L1。
L3 跳过设计
触发条件： 纯后端/API、纯数据处理、基础设施、无 UI 变更 流程： forge-dev 调度器自动判断 → 直接进 forge-eng
自动升级触发器
| 触发器 | 条件 | 动作 |
|---|---|---|
| 迭代不满 | 历史账本显示同一模块 ≥ 3 次反复迭代 | L2 → L1，强制重新设计 |
| 子模块一致性 | 大需求下的子模块 | 必须读父页面 DESIGN.md，检查一致性 |
| 评分过低 | L2 评分 < C | L2 → L1，完整重来 |
设计自检要点
审计和方案设计时逐条检查：
- 层级可答 — 每个页面能回答"用户先看什么、再看什么、最后看什么"；答不出即层级失败。
- 边界情况覆盖 — 超长文本（47 字符的名字）、零结果、网络断开、色盲，每项至少检查一次。
- 减法检查 — 交付前问"如果只能展示 3 样东西，留哪 3 样？"，其余降级或删除。
- 体系一致 — 不孤立评估单个页面；对照前后页面和已有 token 检查一致性。
完整的 12 条设计师认知模式与设计批评话术（背景素养）见 references/create-mode.md。
提问格式与批量策略见
~/.claude/skills/_shared/interaction-protocol.md。设计选项附视觉效果描述。
第0步：定位项目文档 + 门控判定
0.1 定位文档
- 定位 PRD：搜索
{项目目录}/docs/PRD.md或类似文件 - 定位 DESIGN.md：
搜索模式： - {项目目录}/docs/DESIGN.md - {项目目录}/DESIGN.md - {项目目录}/docs/*design*（不区分大小写） - {项目目录}/**/DESIGN*.md - 定位 DESIGN CHANGELOG：模式匹配
*design*changelog*
What ships with it
15 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.
- data/charts.csv 19 KB
- data/colors.csv 32 KB
- data/landing.csv 16 KB
- data/products.csv 57 KB
- data/styles.csv 139 KB
- data/typography.csv 49 KB
- data/ui-reasoning.csv 52 KB
- data/ux-guidelines.csv 18 KB
- README.md 2.1 KB
- references/audit-and-scoring.md 9.2 KB
- references/create-mode.md 8.7 KB
- references/design-template.md 5.2 KB
- scripts/core.py 11 KB runs code
- scripts/design_system.py 43 KB runs code
- scripts/search.py 5.4 KB runs code
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.
- 19d ago Changed · +7 lines · +25 tokens per session 5862b0995bfe
- 2mo ago First seen · 313 lines · 95 tokens per session scan A 4d8330aed096
forge-design is a skill published in the GitHub repository yike-gunshi/forge-skills (13 stars, last pushed 15d ago), licensed MIT. It adds 120 tokens to every session and 4,078 once invoked, about $0.0005 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 2026-08-30.
Other skills, from other repositories
theme-factory
Create new AiderDesk UI themes by defining SCSS color variables, registering theme types, and adding i18n display names. Use when adding a theme, creating a color scheme, customizing appearance, or implementing dark mode and light mode variants.
design-to-code
Mockup-to-component pipeline using Google Stitch, 21st.dev, and Storybook MCP. Accepts a screenshot, a description, or a URL and produces production-ready React components, checking existing Storybook components before generating anything new. Use when implementing UI from a mockup or screenshot. To call the MCP tool…
design-context-extract
Extract design DNA from app screenshots, live URLs, or screen recordings using Google Stitch — color palettes, typography, spacing tokens, component patterns, and motion specs as design-tokens.json or Tailwind config. Use when the user points to a screenshot, URL, or video and asks to extract or audit the design…
design-import
Scaffolds React components from a Claude Design handoff bundle and stops at files on disk: no stories, no tests, no pull request. Use when handed a claude.ai/design URL or a local bundle file; when that same scaffold should carry on through test generation, browser verification and an opened PR, run /ork:design-ship…
multi-surface-render
Multi-surface rendering with json-render — one JSON spec produces React web, Next.js, React Native, Ink terminal UIs, PDFs, emails, Remotion videos, OG images, and 3D scenes. Covers renderer target selection, registry mapping, and platform APIs (renderToBuffer, renderToStream, renderToFile). Use when generating output…
ai-ui-generation
AI-assisted UI generation patterns for json-render, v0.app, Google Stitch, Bolt Cloud, and Cursor workflows. Covers prompt engineering for component and full-stack app generation, review checklists for AI-generated code, design token injection, refactoring for design system conformance, and CI gates for quality…