Token导航 LogoToken导航TokenDH.com
前端设计执行命令github未标认证来源可访问许可证需确认审计通过

design-patterns设计模式

Agent Skill

用于辅助界面设计、视觉规范、排版、配色、布局和交互体验优化。它适合让 Agent 根据产品场景整理页面结构、生成 UI 方案、检查视觉一致性或改进组件层级。使用时需要结合现有品牌、设计系统和用户任务,不应只堆装饰元素;涉及真实页面改动时,应通过截图或浏览器预览检查文本溢出、对齐和响应式表现。

总安装

3,941

周安装

161

GitHub Stars

38,602

下载量

1,275
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:design-patterns(设计模式)
来源仓库:https://github.com/rtk-ai/rtk
仓库路径:skills/design-patterns
安装命令:
npx skills add https://github.com/rtk-ai/rtk --skill design-patterns
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/rtk-ai/rtk --skill design-patterns

简介

用于辅助界面设计、视觉规范、排版、配色、布局和交互体验优化,适合让 Agent 根据产品场景整理页面结构、生成 UI 方案、检查视觉一致性或改进组件层级。

  • 适用于界面设计、视觉规范和交互体验优化等前端设计场景。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装并使用。
  • 使用时需要结合现有品牌、设计系统和用户任务,不应只堆装饰元素。
  • design-patterns 属于前端设计类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

RTK Rust Design Patterns

Patterns that apply to RTK's filter module architecture. Focused on CLI tool patterns, not web/service patterns.

Pattern 1: Newtype (Type Safety)

Use when: wrapping primitive types to prevent misuse (command names, paths, token counts).

// Without Newtype — easy to mix up
fn track(input_tokens: usize, output_tokens: usize) { ... }
track(output_tokens, input_tokens);  // Silent bug!

// With Newtype — compile error on swap
pub struct InputTokens(pub usize);
pub struct OutputTokens(pub usize);
fn track(input: InputTokens, output: OutputTokens) { ... }
track(OutputTokens(100), InputTokens(400));  // Compile error ✅
// Practical RTK example: command name validation
pub struct CommandName(String);
impl CommandName {
    pub fn new(s: &str) -> Result<Self> {
        if s.contains(';') || s.contains('|') || s.contains('`') {
            anyhow::bail!("Invalid command name: shell metacharacters");
        }
        Ok(Self(s.to_string()))
    }
    pub fn as_str(&self) -> &str { &self.0 }
}

Pattern 2: Builder (Complex Configuration)

Use when: a struct has 4+ optional fields, many with defaults.

#[derive(Default)]
pub struct FilterConfig {
    max_lines: Option<usize>,
    strip_ansi: bool,
    show_warnings: bool,
    truncate_at: Option<usize>,
}

impl FilterConfig {
    pub fn new() -> Self { Self::default() }
    pub fn max_lines(mut self, n: usize) -> Self { self.max_lines = Some(n); self }
    pub fn strip_ansi(mut self, v: bool) -> Self { self.strip_ansi = v; self }
    pub fn show_warnings(mut self, v: bool) -> Self { self.show_warnings = v; self }
}

// Usage — readable, no positional arg confusion
let config = FilterConfig::new()
    .max_lines(50)
    .strip_ansi(true)
    .show_warnings(false);

When NOT to use Builder: if the struct has 1-3 fields with obvious meaning. Over-engineering for simple cases.

Pattern 3: State Machine (Parser/Filter Flows)

Use when: parsing multi-section output (test results, build output) where context changes behavior.

// RTK example: pytest output parsing
#[derive(Debug, PartialEq)]
enum ParseState {
    LookingForTests,
    InTestOutput,
    InFailureSummary,
    Done,
}

fn parse_pytest(input: &str) -> String {
    let mut state = ParseState::LookingForTests;
    let mut failures = Vec::new();

    for line in input.lines() {
        match state {
            ParseState::LookingForTests => {
                if line.contains("FAILED") || line.contains("ERROR") {
                    state = ParseState::InFailureSummary;
                    failures.push(line);
                }
            }
            ParseState::InFailureSummary => {
                if line.starts_with("=====") { state = ParseState::Done; }
                else { failures.push(line); }
            }
            ParseState::Done => break,
            _ => {}
        }
    }
    failures.join("\n")
}

Pattern 4: Trait Object (Command Dispatch)

Use when: different command families need the same interface. Avoids massive match arms.

// Define a common interface for filters
pub trait OutputFilter {
    fn filter(&self, input: &str) -> Result<String>;
    fn command_name(&self) -> &str;
}

pub struct GitFilter;
pub struct CargoFilter;

impl OutputFilter for GitFilter {
    fn filter(&self, input: &str) -> Result<String> { filter_git(input) }
    fn command_name(&self) -> &str { "git" }
}

// RTK currently uses match-based dispatch in main.rs (simpler, no dynamic dispatch overhead)
// Trait objects are useful if filter registry becomes dynamic (e.g., TOML-loaded plugins)

Note: RTK's current match dispatch in main.rs is intentional — static dispatch, zero overhead. Only move to trait objects if the match arm count exceeds ~20 commands.

Pattern 5: RAII (Resource Management)

Use when: managing resources that need cleanup (temp files, SQLite connections).

// RTK tee.rs — RAII for temp output files
pub struct TeeFile {
    path: PathBuf,
}

impl TeeFile {
    pub fn create(content: &str) -> Result<Self> {
        let path = tee_path()?;
        fs::write(&path, content)
            .with_context(|| format!("Failed to write tee file: {}", path.display()))?;
        Ok(Self { path })
    }

    pub fn path(&self) -> &Path { &self.path }
}

// No explicit cleanup needed — file persists intentionally (rotation handled separately)
// If cleanup were needed: impl Drop { fn drop(&mut self) { let _ = fs::remove_file(&self.path); } }

Pattern 6: Strategy (Swappable Filter Logic)

Use when: a command has multiple filtering modes (e.g., compact vs. verbose).

pub enum FilterMode {
    Compact,    // Show only failures/errors
    Summary,    // Show counts + top errors
    Full,       // Pass through unchanged
}

pub fn apply_filter(input: &str, mode: FilterMode) -> String {
    match mode {
        FilterMode::Compact => filter_compact(input),
        FilterMode::Summary => filter_summary(input),
        FilterMode::Full => input.to_string(),
    }
}

Pattern 7: Extension Trait (Add Methods to External Types)

Use when: you need to add methods to types you don't own (like &str for RTK-specific parsing).

pub trait RtkStrExt {
    fn is_error_line(&self) -> bool;
    fn is_warning_line(&self) -> bool;
    fn token_count(&self) -> usize;
}

impl RtkStrExt for str {
    fn is_error_line(&self) -> bool {
        self.starts_with("error") || self.contains("[E")
    }
    fn is_warning_line(&self) -> bool {
        self.starts_with("warning")
    }
    fn token_count(&self) -> usize {
        self.split_whitespace().count()
    }
}

// Usage
if line.is_error_line() { ... }
let tokens = output.token_count();

RTK Pattern Selection Guide

SituationPatternAvoid
New *_cmd.rs filter moduleStandard module pattern (see CLAUDE.md)Over-abstracting
4+ optional config fieldsBuilderStruct literal
Multi-phase output parsingState MachineNested if/else
Type-safe wrapper around stringNewtypeRaw String
Adding methods to &strExtension TraitFree functions
Resource with cleanupRAII / DropManual cleanup
Dynamic filter registryTrait ObjectMatch sprawl

Anti-Patterns in RTK Context

// ❌ Generic over-engineering for one command
pub trait Filterable<T: CommandArgs + Send + Sync + 'static> { ... }

// ✅ Just write the function
pub fn filter_git_log(input: &str) -> Result<String> { ... }

// ❌ Singleton registry with global state
static FILTER_REGISTRY: Mutex<HashMap<String, Box<dyn Filter>>> = ...;

// ✅ Match in main.rs — simple, zero overhead, easy to trace

// ❌ Async traits for "future-proofing"
#[async_trait]
pub trait Filter { async fn apply(&self, input: &str) -> Result<String>; }

// ✅ Synchronous — RTK is single-threaded by design
pub trait Filter { fn apply(&self, input: &str) -> Result<String>; }

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

31.88%
按下载量换算406

Claude

31.24%
按下载量换算398

Cursor

17.55%
按下载量换算224

Gemini CLI

10.12%
按下载量换算129

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/rtk-ai/rtk --skill design-patterns 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills