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clear审计异常

bun-devBun DEV 搜索

Agent Skill

bun-dev 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

563

周安装

23

GitHub Stars

26

下载量

180
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

3

许可证

MIT

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:bun-dev(Bun DEV 搜索)
来源仓库:https://github.com/outfitter-dev/agents
仓库路径:skills/bun-dev
安装命令:
npx skills add https://github.com/outfitter-dev/agents --skill bun-dev
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/outfitter-dev/agents --skill bun-dev

简介

用于检索 Bun 开发相关的模式和实践。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。

  • 聚焦原生 API 使用和零依赖架构设计。
  • 涵盖 SQLite、HTTP 服务器和测试等核心模块。
  • 不适合 Node.js 专属模式或跨运行时库开发。
  • 建议结合具体项目需求筛选适用技术方案。bun-dev 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Bun Development

Bun runtime → native APIs → zero-dependency patterns.

<when_to_use>

  • Bun runtime development
  • SQLite database with bun:sqlite
  • HTTP server with Bun.serve
  • Testing with bun:test
  • File operations with Bun.file/Bun.write
  • Shell operations with $ template
  • Password hashing with Bun.password
  • Environment variable handling
  • Building and bundling

NOT for: Node.js-only patterns, cross-runtime libraries, non-Bun projects

</when_to_use>

<runtime_basics>

Package management:

bun install          # Install deps
bun add zod          # Add package
bun remove zod       # Remove package
bun update           # Update all

Script execution:

bun run dev          # Run package.json script
bun run src/index.ts # Execute TypeScript directly
bun --watch index.ts # Watch mode

Testing:

bun test             # All tests
bun test src/        # Directory
bun test --watch     # Watch mode
bun test --coverage  # With coverage

Building:

bun build ./index.ts --outfile dist/bundle.js
bun build ./index.ts --compile --outfile myapp  # Standalone executable

</runtime_basics>

File Operations

<file_operations>

// Read file (lazy, efficient)
const file = Bun.file('./data.json');
if (!(await file.exists())) throw new Error('File not found');

// Read formats
const text = await file.text();
const json = await file.json();
const buffer = await file.arrayBuffer();
const stream = file.stream(); // Large files

// Metadata
console.log(file.size, file.type);

// Write
await Bun.write('./output.txt', 'content');
await Bun.write('./data.json', JSON.stringify(data));
await Bun.write('./blob.txt', new Blob(['data']));

</file_operations>

SQLite (bun:sqlite)

import { Database } from 'bun:sqlite';

const db = new Database('app.db', { create: true, readwrite: true, strict: true });

// Create tables
db.run(`
  CREATE TABLE IF NOT EXISTS users (
    id TEXT PRIMARY KEY,
    email TEXT UNIQUE NOT NULL,
    name TEXT NOT NULL,
    created_at TEXT DEFAULT CURRENT_TIMESTAMP
  )
`);

// Prepared statements (always use these)
const getUser = db.prepare('SELECT * FROM users WHERE id = ?');
const createUser = db.prepare('INSERT INTO users (id, email, name) VALUES (?, ?, ?) RETURNING *');

// Execution
const user = getUser.get('user-123');                    // Single row
const all = db.prepare('SELECT * FROM users').all();     // All rows
db.prepare('DELETE FROM users WHERE id = ?').run('id');  // No return

// Named parameters
const stmt = db.prepare('SELECT * FROM users WHERE email = $email');
stmt.get({ $email: 'alice@example.com' });

// Transactions (atomic, auto-rollback on error)
const transfer = db.transaction((fromId: string, toId: string, amount: number) => {
  db.run('UPDATE accounts SET balance = balance - ? WHERE id = ?', [amount, fromId]);
  db.run('UPDATE accounts SET balance = balance + ? WHERE id = ?', [amount, toId]);
});
transfer('alice', 'bob', 100);

db.close(); // When done

See sqlite-patterns.md for migrations, pooling, repository pattern.

Password Hashing

// Hash (argon2id recommended)
const hash = await Bun.password.hash('password123', {
  algorithm: 'argon2id',
  memoryCost: 65536,  // 64 MB
  timeCost: 3
});

// Or bcrypt
const bcryptHash = await Bun.password.hash('password123', {
  algorithm: 'bcrypt',
  cost: 12
});

// Verify
const isValid = await Bun.password.verify('password123', hash);
if (!isValid) throw new Error('Invalid password');

Auth flow example:

app.post('/auth/register', zValidator('json', RegisterSchema), async (c) => {
  const { email, password } = c.req.valid('json');
  const db = c.get('db');

  if (db.prepare('SELECT id FROM users WHERE email = ?').get(email)) {
    throw new HTTPException(409, { message: 'Email already registered' });
  }

  const hashedPassword = await Bun.password.hash(password, { algorithm: 'argon2id' });
  const user = db.prepare(`
    INSERT INTO users (id, email, password) VALUES (?, ?, ?) RETURNING id, email
  `).get(crypto.randomUUID(), email, hashedPassword);

  return c.json({ user }, 201);
});

HTTP Server

<http_server>

Bun.serve({
  port: 3000,
  fetch(req) {
    const url = new URL(req.url);
    if (url.pathname === '/') return new Response('Hello');
    if (url.pathname === '/json') return Response.json({ ok: true });
    return new Response('Not found', { status: 404 });
  },
  error(err) {
    return new Response(`Error: ${err.message}`, { status: 500 });
  }
});

With Hono (recommended for APIs):

import { Hono } from 'hono';

const app = new Hono()
  .get('/', (c) => c.text('Hello'))
  .get('/json', (c) => c.json({ ok: true }));

Bun.serve({ port: 3000, fetch: app.fetch });

See server-patterns.md for routing, middleware, file serving, streaming.

</http_server>

WebSocket

import type { ServerWebSocket } from 'bun';

type WsData = { userId: string };

Bun.serve<WsData>({
  port: 3000,
  fetch(req, server) {
    const url = new URL(req.url);
    if (url.pathname === '/ws') {
      const userId = url.searchParams.get('userId') || 'anon';
      return server.upgrade(req, { data: { userId } }) ? undefined
        : new Response('Upgrade failed', { status: 400 });
    }
    return new Response('Hello');
  },
  websocket: {
    open(ws: ServerWebSocket<WsData>) {
      ws.subscribe('chat');
      ws.send(JSON.stringify({ type: 'connected' }));
    },
    message(ws: ServerWebSocket<WsData>, msg: string | Buffer) {
      ws.publish('chat', msg);
    },
    close(ws: ServerWebSocket<WsData>) {
      ws.unsubscribe('chat');
    }
  }
});

See server-patterns.md for client tracking, rooms, reconnection.

Shell Operations

import { $ } from 'bun';

// Run commands
const result = await $`ls -la`;
console.log(result.text());

// Variables (auto-escaped)
const dir = './src';
await $`find ${dir} -name "*.ts"`;

// Check exit code
const { exitCode } = await $`npm test`.nothrow();
if (exitCode !== 0) console.error('Tests failed');

// Spawn process
const proc = Bun.spawn(['ls', '-la']);
await proc.exited;

// Capture output
const proc2 = Bun.spawn(['echo', 'Hello'], { stdout: 'pipe' });
const output = await new Response(proc2.stdout).text();

Testing (bun:test)

import { describe, test, expect, beforeEach, afterEach } from 'bun:test';

describe('feature', () => {
  let db: Database;

  beforeEach(() => { db = new Database(':memory:'); });
  afterEach(() => { db.close(); });

  test('behavior', () => {
    expect(result).toBe(expected);
    expect(arr).toContain(item);
    expect(fn).toThrow();
    expect(obj).toEqual({ foo: 'bar' });
  });

  test('async', async () => {
    const result = await asyncFn();
    expect(result).toBeDefined();
  });

  test.todo('pending feature');
  test.skip('temporarily disabled');
});
bun test                    # All tests
bun test src/api.test.ts    # Specific file
bun test --watch            # Watch mode
bun test --coverage         # With coverage

See testing.md for assertions, mocking, snapshots, best practices.

Environment Variables

// Access
console.log(Bun.env.NODE_ENV);
console.log(Bun.env.DATABASE_URL);

// Zod validation
import { z } from 'zod';

const EnvSchema = z.object({
  NODE_ENV: z.enum(['development', 'production', 'test']).default('development'),
  DATABASE_URL: z.string(),
  PORT: z.coerce.number().int().positive().default(3000),
  API_KEY: z.string().min(32)
});

export const env = EnvSchema.parse(Bun.env);

Bun auto-loads .env, .env.local, .env.production.

Performance Utilities

// High-resolution timing
const start = Bun.nanoseconds();
await doWork();
console.log(`Took ${(Bun.nanoseconds() - start) / 1_000_000}ms`);

// Hashing
const hash = Bun.hash(data);
const crc32 = Bun.hash.crc32(data);
const sha256 = Bun.CryptoHasher.hash('sha256', data);

// Sleep
await Bun.sleep(1000);

// Memory
const { rss, heapUsed } = process.memoryUsage();
console.log('RSS:', rss / 1024 / 1024, 'MB');

Building & Bundling

# Production bundle
bun build ./index.ts --outfile dist/bundle.js --minify --sourcemap

# External deps
bun build ./index.ts --outfile dist/bundle.js --external hono --external zod

# Standalone executable
bun build ./index.ts --compile --outfile myapp

# Cross-compile
bun build ./index.ts --compile --target=bun-linux-x64 --outfile myapp-linux
bun build ./index.ts --compile --target=bun-darwin-arm64 --outfile myapp-macos
bun build ./index.ts --compile --target=bun-windows-x64 --outfile myapp.exe

ALWAYS:

  • Use Bun APIs when available (faster, native)
  • Prepared statements for database queries
  • Transactions for multi-statement operations
  • argon2id for password hashing
  • Validate environment variables at startup
  • Close database connections when done

NEVER:

  • String interpolation in SQL (use parameters)
  • Plaintext passwords
  • Ignore async disposal cleanup
  • Deprecated Node.js APIs when Bun native exists

PREFER:

  • Bun.file over fs.readFile
  • Bun.write over fs.writeFile
  • bun:sqlite over external SQLite libraries
  • Bun.password over bcrypt/argon2 packages
  • $ shell template over child_process

Examples:

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

04

需要参考平台分布和安装热度时

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

补充不同宿主或平台的使用分布数据

能力 5

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

Claude Code

28.88%
按下载量换算52

Cursor

21.27%
按下载量换算38

windsurf

16.26%
按下载量换算29

trae

13.96%
按下载量换算25

OpenCode

8.08%
按下载量换算15

Codex

3.25%
按下载量换算6

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

可疑

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。

来源信息

继续浏览同类 Skills