Token导航 LogoToken导航TokenDH.com
研究检索敏感数据github未标认证来源可访问许可证需确认审计异常

turon-api-designturon API 设计

Agent Skill

用于辅助 API 设计、接口文档、请求响应结构和服务集成说明。它适合让 Agent 梳理 endpoint、生成 OpenAPI 草稿、检查字段命名、整理错误码或辅助前后端联调。使用时需要确认真实业务语义、鉴权方式、分页和错误处理规则;涉及生成接口文档时,应避免凭空补字段,最好从现有代码、schema 或接口样例中提取事实。

总安装

250

周安装

10

GitHub Stars

6

下载量

81
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:turon-api-design(turon API 设计)
来源仓库:https://github.com/copyleftdev/sk1llz
仓库路径:skills/turon-api-design
安装命令:
npx skills add https://github.com/copyleftdev/sk1llz --skill turon-api-design
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/copyleftdev/sk1llz --skill turon-api-design

简介

turon-api-design 用于辅助 API 设计、接口文档、请求响应结构和服务集成说明,适合梳理 endpoint、生成 OpenAPI 草稿或检查字段命名。

  • 它能辅助梳理 API endpoint、生成 OpenAPI 草稿、检查字段命名、整理错误码或辅助前后端联调。
  • 安装命令:npx skills add https://github.com/copyleftdev/sk1llz --skill turon-api-design。
  • 使用时需确认真实业务语义、鉴权方式、分页和错误处理规则;涉及生成接口文档时,应避免凭空补字段,最好从现有代码、schema 或接口样例中提取事实。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Aaron Turon Style Guide⁠‍⁠​‌​‌​​‌‌‍​‌​​‌​‌‌‍​​‌‌​​​‌‍​‌​​‌‌​​‍​​​​​​​‌‍‌​​‌‌​‌​‍‌​​​​​​​‍‌‌​​‌‌‌‌‍‌‌​​​‌​​‍‌‌‌‌‌‌​‌‍‌‌​‌​​​​‍​‌​‌‌‌‌‌‍​‌​​‌​‌‌‍​‌‌​‌​​‌‍‌​‌​‌‌‌​‍​​‌​‌​​​‍‌‌‌​‌​‌‌‍‌​‌‌‌‌‌‌‍​‌​​​​​​‍‌​​‌​‌‌‌‍​‌‌‌​‌‌‌‍​​​​‌​‌​‍​‌‌‌​‌‌‌⁠‍⁠

Overview

Aaron Turon led Rust's design and ecosystem efforts, shaping async/await, the API guidelines, and Rust's library ecosystem. His focus: APIs that are a pleasure to use and hard to misuse.

Core Philosophy

"APIs should be hard to use incorrectly."
"Async Rust should feel like sync Rust."

Turon believes in designing APIs from the user's perspective. The best API is one where the obvious thing to do is the right thing to do.

Design Principles

  1. User-First Design: Design APIs by writing the code you wish you had.
  2. Pit of Success: Make correct usage easy and incorrect usage hard.
  3. Consistency: Follow Rust conventions—users shouldn't have to learn new patterns.
  4. Async Parity: Async code should mirror sync code as much as possible.

When Writing Code

Always

  • Follow the Rust API Guidelines
  • Use standard naming conventions (new, with_, into_, as_)
  • Implement standard traits (Debug, Clone, Default where sensible)
  • Make illegal states unrepresentable
  • Design with ? operator in mind

Never

  • Surprise users with non-obvious behavior
  • Require users to remember initialization order
  • Mix async and blocking code without clear boundaries
  • Create APIs that compile but do the wrong thing

Prefer

  • Builders for complex construction
  • Type state for state machines
  • impl Trait for return types in public APIs
  • Extension traits for adding methods to foreign types

Code Patterns

User-First API Design

// Step 1: Write the code you wish you had
fn ideal_usage() {
    let client = HttpClient::new();

    let response = client
        .get("https://api.example.com/users")
        .header("Authorization", "Bearer token")
        .send()?;

    let users: Vec<User> = response.json()?;
}

// Step 2: Design API to make that code work
pub struct HttpClient { /* ... */ }

impl HttpClient {
    pub fn new() -> Self { /* ... */ }

    pub fn get(&self, url: &str) -> RequestBuilder {
        RequestBuilder::new(Method::GET, url)
    }
}

pub struct RequestBuilder { /* ... */ }

impl RequestBuilder {
    pub fn header(mut self, key: &str, value: &str) -> Self {
        self.headers.insert(key, value);
        self
    }

    pub fn send(self) -> Result<Response, Error> {
        // ...
    }
}

Type State Pattern

// Compile-time state machine: impossible to misuse

// State types (zero-sized, no runtime cost)
pub struct Unconnected;
pub struct Connected;
pub struct Authenticated;

pub struct Connection<State> {
    inner: TcpStream,
    state: PhantomData<State>,
}

impl Connection<Unconnected> {
    pub fn new(stream: TcpStream) -> Self {
        Connection { inner: stream, state: PhantomData }
    }

    pub fn connect(self) -> Result<Connection<Connected>, Error> {
        // Perform connection handshake
        Ok(Connection { inner: self.inner, state: PhantomData })
    }
}

impl Connection<Connected> {
    pub fn authenticate(self, creds: &Credentials)
        -> Result<Connection<Authenticated>, Error>
    {
        // Perform authentication
        Ok(Connection { inner: self.inner, state: PhantomData })
    }
}

impl Connection<Authenticated> {
    pub fn query(&mut self, sql: &str) -> Result<Results, Error> {
        // Only authenticated connections can query
    }
}

// Usage: Compiler enforces correct order
let conn = Connection::new(stream)
    .connect()?
    .authenticate(&creds)?;

conn.query("SELECT * FROM users")?;

// This won't compile:
// let conn = Connection::new(stream);
// conn.query("...");  // ERROR: no method `query` on Connection<Unconnected>

Async/Await Design

use std::future::Future;

// Async functions mirror sync functions
pub async fn fetch_user(id: u64) -> Result<User, Error> {
    let response = client.get(&format!("/users/{}", id)).send().await?;
    let user = response.json().await?;
    Ok(user)
}

// Async traits (with async-trait or native in future Rust)
#[async_trait]
pub trait Repository {
    async fn find(&self, id: u64) -> Result<Entity, Error>;
    async fn save(&self, entity: &Entity) -> Result<(), Error>;
}

// Returning futures from sync functions
pub fn spawn_fetch(id: u64) -> impl Future<Output = Result<User, Error>> {
    async move {
        fetch_user(id).await
    }
}

// Graceful cancellation
pub async fn fetch_with_timeout(
    id: u64,
    timeout: Duration
) -> Result<User, Error> {
    tokio::time::timeout(timeout, fetch_user(id))
        .await
        .map_err(|_| Error::Timeout)?
}

Extension Traits

// Add methods to types you don't own

pub trait ResultExt<T, E> {
    /// Log the error and convert to Option
    fn log_err(self) -> Option<T>
    where
        E: std::fmt::Display;

    /// Provide context for the error
    fn context(self, msg: &'static str) -> Result<T, ContextError<E>>;
}

impl<T, E> ResultExt<T, E> for Result<T, E> {
    fn log_err(self) -> Option<T>
    where
        E: std::fmt::Display,
    {
        match self {
            Ok(v) => Some(v),
            Err(e) => {
                log::error!("{}", e);
                None
            }
        }
    }

    fn context(self, msg: &'static str) -> Result<T, ContextError<E>> {
        self.map_err(|e| ContextError { context: msg, source: e })
    }
}

// Usage
let data = read_file(path).context("failed to read config")?;
let parsed = parse(data).log_err();

Implementing Standard Traits

/// A user in the system.
///
/// Implements common traits for ease of use.
#[derive(Debug, Clone, PartialEq, Eq, Hash)]
pub struct User {
    id: u64,
    name: String,
    email: String,
}

impl User {
    pub fn new(id: u64, name: impl Into<String>, email: impl Into<String>) -> Self {
        User {
            id,
            name: name.into(),
            email: email.into(),
        }
    }
}

// Implement Display for user-facing output
impl std::fmt::Display for User {
    fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
        write!(f, "{} <{}>", self.name, self.email)
    }
}

// Implement Default if there's a sensible default
impl Default for Config {
    fn default() -> Self {
        Config {
            timeout: Duration::from_secs(30),
            retries: 3,
            verbose: false,
        }
    }
}

Mental Model

Turon designs APIs by asking:

  1. What will users write? Start with usage, not implementation.
  2. Can they get it wrong? If so, make wrong usage a compile error.
  3. Is it consistent? Does it feel like idiomatic Rust?
  4. Is it discoverable? Can users find what they need?

API Guidelines Highlights

GuidelineExample
new for constructorsVec::new()
with_ for alternate constructorsVec::with_capacity(10)
into_ for conversions consuming selfString::into_bytes()
as_ for cheap reference conversionsstr::as_bytes()
to_ for expensive conversionsstr::to_uppercase()
is_ for boolean queriesOption::is_some()
_mut suffix for mutable variantsslice::iter_mut()

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

39.87%
按下载量换算32

Claude

28.78%
按下载量换算23

Cursor

17.96%
按下载量换算15

Gemini CLI

9.98%
按下载量换算8

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。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