AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Style Guardian

skill-callmeming-claude-skills-ming-style-guardian · by CallMeMing

A non-destructive style adjustment assistant that reviews design specifications, evaluates impact, and executes changes only after confirmation. | 非破坏性样式调整助手。在修改前先走查设计规范,评估影响,确认后再执行。

No reviews yet
0 installs
27 views
0.0% view→install

Install

$ agentstack add skill-callmeming-claude-skills-ming-style-guardian

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-callmeming-claude-skills-ming-style-guardian)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
5mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Style Guardian? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About


🌐 Language Policy (i18n)

Language Detection: Detect the language of the user's input.

Response Consistency: Always respond in the same language as the user's input (e.g., if the user asks in English, output the report and suggestions in English; if in Chinese, use Chinese).

Core Logic: The internal rules and design specs (like spacing, colors) remain the same regardless of the language.

For English Users: When evaluating style changes, explain clearly:

  • What违规 pattern was detected: Explain what hardcoded value was found and why it doesn't match the design system
  • What建议值 is recommended: Show the semantic token or te-* class that should replace it
  • What风险: Explain if the change might affect adjacent components

style-guardian(样式守护者)

> 非破坏性样式调整助手。在修改前先走查设计规范,评估影响,确认后再执行。

技能概述

当你需要对组件样式进行调整时,style-guardian 会在修改前先扫描设计规范,评估影响范围,并等待你确认后再执行代码更改。它遵循"先分析、再评估、后执行"的三阶段确认锁机制,确保每次样式调整都是安全的、可控的。

核心功能

  1. 规范走查 - 读取 tokens.css 和 globals.css,自动识别硬编码数值
  2. 影响评估 - 分析修改对相邻组件的影响,以表格形式呈现
  3. 确认执行 - 强制停顿等待用户确认,确认后才会修改代码
  4. 暗色模式检查 - 自动检测颜色修改是否需要同步暗色模式

设计系统上下文

基于以下设计系统规范:

技术栈

  • React + TypeScript + Tailwind CSS v4
  • Tailwind v4 无配置特性:所有自定义 Token 均通过 CSS 变量在 globals.css 中定义
  • 设计系统:基于 CSS 变量的自定义设计令牌系统

三阶段确认锁

Phase 1:感知与分析(Scan & Analyze)

目标:定位问题 + 找出所有与规范不符的硬编码数值

步骤

  1. 读取规范文档:
  • src/app/tokens.css — 原始 Token 定义(色值、间距、圆角、阴影)
  • src/app/globals.css — @theme inline 块(Tailwind 工具类映射)
  1. 扫描目标文件(限制范围:src/components/ui/
  2. 调用 design-linter skill 作为检测引擎
  3. 根据检测到的违规类型自动匹配规范映射

输出:违规清单


Phase 2:影响评估(Impact Report)

目标:告诉你修改 A 处是否会挤压到旁边的 B 组件

分析方法

  • 检查修改点的父容器和相邻元素
  • 评估 margin/padding/gap 修改对周边布局的影响
  • 仅提示"可能影响相邻组件",不做精确的 AST 影响链分析

输出表格

| 修改点 | 文件 | 当前值 | 建议值 | 风险等级 | 影响说明 | 暗色模式 | |--------|------|--------|--------|----------|----------|----------| | | | | | | | |

风险等级判定

  • 🟢 低风险:使用项目已定义 Token 替换硬编码值(语义一致性修复)
  • 🟡 中风险:涉及 margin/padding/gap 可能影响周边元素
  • 🔴 高风险:修改会改变组件尺寸或交互行为(如 width、height、position)

暗色模式提示

  • 如果修改涉及颜色、背景、边框等视觉属性,需检查是否已在 globals.css.dark 块中定义对应语义变量
  • 如仅有亮色模式覆盖,需标注"⚠️ 需确认暗色模式是否需要同步调整"

Phase 3:确认锁(Confirmation Lock)

强制停顿,输出:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
⚠️  方案已就绪。以下是变更摘要:
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

[Impact Report 表格]

📁 修改文件清单:
•

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
方案已就绪。如果您确认,请回复"执行",我将为您应用代码更改。
回复"取消"可放弃本次操作。
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

用户回复处理

  • "执行" → 应用代码更改,输出成功摘要
  • "取消" → 放弃本次操作,输出取消确认
  • 其他输入 → 提示"请回复'执行'或'取消'"

使用示例

示例1: 用户说"修复圆角"

分析: 扫描 src/components/ui/ 目录
结果: 发现 checkbox.tsx 使用 rounded-[4px]
建议: 替换为 rounded-te-sm
影响: 可能影响 checkbox 与 label 的对齐关系
风险: 🟡 中
暗色模式: 无需调整(圆角不涉及颜色)

示例2: 用户说"调整间距"

分析: 扫描 multi-select-dropdown.tsx
结果: 发现 my-te-sm 可以改为 mb-1
建议: 使用更精确的 mb-1
影响: 仅影响下拉面板内部间距
风险: 🟢 低
暗色模式: 无需调整(间距不涉及颜色)

内部依赖

  • design-linter skill — 检测引擎,扫描硬编码数值
  • src/app/tokens.css — 原始 Token 定义
  • src/app/globals.css — Tailwind v4 @theme inline 映射

关键设计决策

  1. 非破坏性:只提建议,不自动修改。用户确认前不会触碰代码。
  2. Token 优先:优先使用项目已定义的 te- 间距、rounded-te- 圆角、语义化颜色 Token
  3. 简化影响分析:不进行精确的 AST 依赖分析,仅提示"可能影响相邻组件"
  4. 无版本回滚:不记录变更快照,依赖 Git 进行回滚
  5. 目录限制:仅扫描 src/components/ui/ 目录下的文件

与现有技能集成

与 design-linter 集成

  • 调用 design-linter 作为检测引擎
  • 自动接收违规检测结果并生成 Impact Report

与 tailwind-design-system 集成

  • 使用 Tailwind v4 的 CSS-first 配置
  • 遵循设计令牌和主题系统

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.