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/xrensiu/claude-code-forge/bespoke-design-systemnpx skills add XRenSiu/claude-code-forge --skill bespoke-design-systemgit clone --depth 1 https://github.com/XRenSiu/claude-code-forgeWrote 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/xrensiu/claude-code-forge/bespoke-design-system)<a href="https://agentmods.dev/skills/xrensiu/claude-code-forge/bespoke-design-system"><img src="https://agentmods.dev/badge/skills/xrensiu/claude-code-forge/bespoke-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 | $0.00239 | $0.10615 |
| Opus 5 | $0.00120 | $0.05307 |
| Sonnet 5 | $0.00048 | $0.02123 |
| Haiku 4.5 | $0.00024 | $0.01061 |
Grade A, and why
bespoke-design-system 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 5d 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 — 533 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Bespoke Design System — 专属设计系统生成器
你被调用来处理 $ARGUMENTS:可能是为一个产品需求生成专属 DESIGN.md(主流程,99% 调用),也可能是维护规则库(增量拆解新素材、批量导入集合、沉淀高频 adaptation、重建关系图)。
开始时告诉用户:
- 主流程:" I'm using the bespoke-design-system skill to generate a tailored DESIGN.md for your product. "
- 维护流程:" I'm using the bespoke-design-system skill in maintenance mode to <update / import / extract / consolidate>. "
一、本质(必读,每次都要在心里过一遍)
把任意一组 DESIGN.md 素材里 " 做对了的设计判断 " 形式化成可计算、可组合、可演化的设计语法,然后用这套语法为每个新用户做一次新的设计判断,且每个判断都必须能回答 " 为什么对这个用户对 "。
产品诚实声明(v4 新增)
本 skill 产出的是带论证的 DESIGN.md 初稿,不是成品。
它能做到的:
- 内在协调性(配色/排版/间距的数学关系正确)— 完全可量化
- 语境贴合性(匹配 archetype 和 Kansei 画像)— 离散判据 + 半量化
- 不抄袭(不是某现有系统换个强调色)— neighbor 独特性带可量化下限(v1.11.0)
- 不平庸(有一个能说清的 POV / 签名动作,不是品类默认值堆叠)— taste-critic 可判的平庸下限(v1.12.0)
⚠️ v1.12.0 起不再声称"不出格"。旧版用"不会跑出已知好设计的范围"当卖点,恰恰是把下限焊成天花板、逼出平庸的病根(详见
bespoke-design/skill-issue-2026-06-18.md)。现在的定位是在协调的前提下争取独特,而不是"安全地待在已知范围内"。
它做不到的:
- 判断"这份设计有大师级品味" — taste-critic 能判"没有观点"(下限),判不了"有伟大品味"(上限)。学界共识:tacit knowledge 无法形式化(Polanyi 1966;CHI 2024)
- 替代设计师的最终判断 — 品味这一关必须由人完成
定位上,它替代的是"用户硬选 5 个固定方向"或"硬选别人的 DESIGN.md"这个低质量起点,不是替代设计师。详见 references/tacit-knowledge-boundary.md。
五条铁律
- 双向 Rationale 对称:拆解时怎么提取 rationale,生成时就怎么产出 rationale。生成时的 rationale 必须三段式:
inheritance(来源)+adaptation(调适)+justification(协同)。 - 一次性追问铁律(interactive):所有澄清问题必须在 B1a 一次性问完,硬上限 7 条,超过则强制 B1a 重新规划聚焦。B2 之后绝不向用户追问任何信息——不管是 B3 冲突、B4 决策、B5 闸门失败,都不许打断用户。
- 品味需自审(v4 新增 / v1.12.0 加 taste-critic):B5 的 6 项 check 全过 ≠ 这份 DESIGN.md 有品味。taste-critic 把"很普通"变成可判的平庸下限,但"有大师级品味"仍是上限、需人审。B6 的 negotiation-summary 必须明确告知用户"品味终审需由人完成"。引用
references/tacit-knowledge-boundary.md。 - 不产生 AI slop:紫渐变 + Inter + 圆角卡片那一套是 anti-pattern。每个决策都要在
references/anti-slop-blacklist.md上过一遍。 - 模式选择交给用户:模式是用户对 " 我愿意花多少时间换多少精度 " 的偏好,不是技术问题。skill 不替用户决定。
二、触发分流(第一步必做)
解析 $ARGUMENTS 与最近的用户消息,判断走哪条主路径:
主流程:生成 DESIGN.md
触发条件:用户描述产品需求并希望得到设计系统、调性方案、UI 风格指南、设计 token、设计基底等。即使用户表述很模糊(例如 " 我想做个八字 SaaS"、" 给我做个设计调性 "),只要意图是 " 获得设计系统 ",都走这里。
What ships with it
60 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.
- checks/_shared/archetype_rules.json 14 KB
- checks/_shared/color_math.py 9.7 KB runs code
- checks/_shared/scale_math.py 6.1 KB runs code
- checks/archetype_check.py 17 KB runs code
- checks/coherence_check.py 21 KB runs code
- checks/kansei_coverage_check.py 9.7 KB runs code
- checks/neighbor_check.py 11 KB runs code
- examples/_README.md 3.5 KB
- examples/code-review-saas-v1.0.0/inferred-fields.yaml 2.7 KB
- examples/code-review-saas-v1.0.0/input.md 240 B
- examples/code-review-saas-v1.0.0/issues-found.md 11 KB
- examples/code-review-saas-v1.0.0/mode.txt 5 B
- examples/code-review-saas-v1.0.0/negotiation-summary.md 4.7 KB
- examples/code-review-saas-v1.0.0/output-design.md 17 KB
- examples/code-review-saas-v1.0.0/provenance.yaml 15 KB
- examples/code-review-saas/check-report.json 7.3 KB
- examples/code-review-saas/inferred-fields.yaml 2.7 KB
- examples/code-review-saas/input.md 240 B
- examples/code-review-saas/mode.txt 5 B
- examples/code-review-saas/negotiation-summary.md 6.3 KB
- examples/code-review-saas/output-design.md 17 KB
- examples/code-review-saas/provenance.yaml 18 KB
- grammar/graph/rules_graph.json 218 KB
- grammar/meta/adaptation-stats.json 44 KB
- grammar/meta/defaults.yaml 5.0 KB
- grammar/meta/neighbor-corpus.json 90 KB
- grammar/meta/provenance-index.json 53 KB
- grammar/meta/source-registry.json 83 KB
- grammar/meta/version.json 3.2 KB
- grammar/rationale/.gitkeep 336 B
- grammar/rationale/airbnb.md 13 KB
- grammar/rationale/airtable.md 2.9 KB
- grammar/rationale/ant.md 1.9 KB
- grammar/rationale/apple.md 11 KB
- grammar/rationale/arc.md 3.4 KB
- grammar/rationale/binance.md 4.4 KB
- grammar/rationale/bmw.md 4.7 KB
- grammar/rationale/brutalism.md 6.6 KB
- grammar/rationale/bugatti.md 4.6 KB
- grammar/rationale/cal.md 4.1 KB
- grammar/rationale/canva.md 3.4 KB
- grammar/rationale/claude.md 12 KB
- grammar/rationale/clickhouse.md 3.7 KB
- grammar/rationale/cohere.md 4.5 KB
- grammar/rationale/coinbase.md 3.1 KB
- grammar/rationale/composio.md 3.7 KB
- grammar/rationale/cursor.md 3.3 KB
- grammar/rationale/discord.md 10 KB
- grammar/rationale/duolingo.md 11 KB
- grammar/rationale/elevenlabs.md 4.1 KB
- grammar/rationale/figma.md 8.6 KB
- grammar/rationale/framer.md 4.0 KB
- grammar/rationale/github.md 4.1 KB
- grammar/rationale/huggingface.md 4.0 KB
- grammar/rationale/intercom.md 4.0 KB
- grammar/rationale/kami.md 4.4 KB
- grammar/rationale/linear-app.md 9.5 KB
- grammar/rationale/lovable.md 4.8 KB
- grammar/rationale/mastercard.md 5.0 KB
- grammar/rationale/mintlify.md 9.8 KB
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.
- 5d ago First seen · 533 lines · 239 tokens per session scan A 2e379d95db4c
bespoke-design-system is a skill published in the GitHub repository XRenSiu/claude-code-forge (2 stars, last pushed today), licensed MIT. It adds 239 tokens to every session and 10,615 once invoked, about $0.0012 per session on Opus 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 2026-08-31.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…
chronicle
Analyze Copilot session history for standup reports, usage tips, session search, and session reindexing. Use when the user asks for a standup, daily summary, usage tips, workflow recommendations, wants to search or find past sessions by keyword/file/PR, wants to reindex their session store, or asks about deleting…