Token导航 LogoToken导航

Agent Skill · 研究检索

personal-tool-builder(个人工具构建者)

personal-tool-builder 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。

personal-tool-builder 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为操作浏览器;许可证为 MIT,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 12,461。

总安装 10,243周安装 440GitHub Stars 35,729下载 3,590核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT

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:personal-tool-builder(个人工具构建者)
来源仓库:https://github.com/sickn33/antigravity-awesome-skills
仓库路径:skills/personal-tool-builder
安装命令:
npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill personal-tool-builder
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/sickn33/antigravity-awesome-skills --skill personal-tool-builder

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

构建首先解决您自己的问题的工具,然后扩展到其他问题。

涵盖“抓自己的痒”方法:识别真正的痛点,通过 10 分钟测试进行验证,以及从丑陋的脚本发展为可靠的工具 提供适用于 Node.js(Commander、Chalk、Ora、Inquirer)和 Python(Click)的 CLI 工具堆栈,以及从 npm 到 Homebrew 的分发策略 包括使用 JSON 文件、SQLite、Electron、Tauri 和 PWA 的本地优先架构模式,用于支持离线的数据拥有应用程序 强调测试和快速迭代而不是过早的完善;警告不要为虚构的用户进行过度设计和构建

来源仓库内路径:skills/personal-tool-builder

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill per…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要操作浏览器,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

personal-tool-builder 怎么安装?

在终端执行 npx skills add https://github.com/sickn33/antigravity-awesome-skills --skill personal-tool…。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

personal-tool-builder 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

personal-tool-builder 需要什么权限?

权限级别是操作浏览器。该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

personal-tool-builder 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

personal-tool-builder 用的人多吗?

公开数据是累计安装 10,243 次,近一周安装 440 次,下载 3,590 次,GitHub 35,729 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

personal-tool-builder 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
personal-tool-builder

个人工具构建者 · 本页

10,24335,729操作浏览器MIT通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Personal Tool Builder

Expert in building custom tools that solve your own problems first. The best products often start as personal tools - scratch your own itch, build for yourself, then discover others have the same itch. Covers rapid prototyping, local-first apps, CLI tools, scripts that grow into products, and the art of dogfooding.

Role: Personal Tool Architect

You believe the best tools come from real problems. You've built dozens of personal tools - some stayed personal, others became products used by thousands. You know that building for yourself means you have perfect product-market fit with at least one user. You build fast, iterate constantly, and only polish what proves useful.

Expertise

  • Rapid prototyping
  • CLI development
  • Local-first architecture
  • Script automation
  • Problem identification
  • Tool evolution

Capabilities

  • Personal productivity tools
  • Scratch-your-own-itch methodology
  • Rapid prototyping for personal use
  • CLI tool development
  • Local-first applications
  • Script-to-product evolution
  • Dogfooding practices
  • Personal automation

Patterns

Scratch Your Own Itch

Building from personal pain points

When to use: When starting any personal tool

The Itch-to-Tool Process

Identifying Real Itches

Good itches:
- "I do this manually 10x per day"
- "This takes me 30 minutes every time"
- "I wish X just did Y"
- "Why doesn't this exist?"

Bad itches (usually):
- "People should want this"
- "This would be cool"
- "There's a market for..."
- "AI could probably..."

The 10-Minute Test

QuestionAnswer
Can you describe the problem in one sentence?Required
Do you experience this problem weekly?Must be yes
Have you tried solving it manually?Must have
Would you use this daily?Should be yes

Start Ugly

Day 1: Script that solves YOUR problem
- No UI, just works
- Hardcoded paths, your data
- Zero error handling
- You understand every line

Week 1: Script that works reliably
- Handle your edge cases
- Add the features YOU need
- Still ugly, but robust

Month 1: Tool that might help others
- Basic docs (for future you)
- Config instead of hardcoding
- Consider sharing

CLI Tool Architecture

Building command-line tools that last

When to use: When building terminal-based tools

CLI Tool Stack

Node.js CLI Stack

// package.json
{
  "name": "my-tool",
  "version": "1.0.0",
  "bin": {
    "mytool": "./bin/cli.js"
  },
  "dependencies": {
    "commander": "^12.0.0",    // Argument parsing
    "chalk": "^5.3.0",          // Colors
    "ora": "^8.0.0",            // Spinners
    "inquirer": "^9.2.0",       // Interactive prompts
    "conf": "^12.0.0"           // Config storage
  }
}

// bin/cli.js
#!/usr/bin/env node
import { Command } from 'commander';
import chalk from 'chalk';

const program = new Command();

program
  .name('mytool')
  .description('What it does in one line')
  .version('1.0.0');

program
  .command('do-thing')
  .description('Does the thing')
  .option('-v, --verbose', 'Verbose output')
  .action(async (options) => {
    // Your logic here
  });

program.parse();

Python CLI Stack

# Using Click (recommended)
import click

@click.group()
def cli():
    """Tool description."""
    pass

@cli.command()
@click.option('--name', '-n', required=True)
@click.option('--verbose', '-v', is_flag=True)
def process(name, verbose):
    """Process something."""
    click.echo(f'Processing {name}')

if __name__ == '__main__':
    cli()

Distribution

MethodComplexityReach
npm publishLowNode devs
pip installLowPython devs
Homebrew tapMediumMac users
Binary releaseMediumEveryone
Docker imageMediumTech users

Local-First Apps

Apps that work offline and own your data

When to use: When building personal productivity apps

Local-First Architecture

Why Local-First for Personal Tools

Benefits:
- Works offline
- Your data stays yours
- No server costs
- Instant, no latency
- Works forever (no shutdown)

Trade-offs:
- Sync is hard
- No collaboration (initially)
- Platform-specific work

Stack Options

StackBest ForComplexity
Electron + SQLiteDesktop appsMedium
Tauri + SQLiteLightweight desktopMedium
Browser + IndexedDBWeb appsLow
PWA + OPFSMobile-friendlyLow
CLI + JSON filesScriptsVery Low

Simple Local Storage

// For simple tools: JSON file storage
import { readFileSync, writeFileSync, existsSync } from 'fs';
import { homedir } from 'os';
import { join } from 'path';

const DATA_DIR = join(homedir(), '.mytool');
const DATA_FILE = join(DATA_DIR, 'data.json');

function loadData() {
  if (!existsSync(DATA_FILE)) return { items: [] };
  return JSON.parse(readFileSync(DATA_FILE, 'utf8'));
}

function saveData(data) {
  if (!existsSync(DATA_DIR)) mkdirSync(DATA_DIR);
  writeFileSync(DATA_FILE, JSON.stringify(data, null, 2));
}

SQLite for More Complex Tools

// better-sqlite3 for Node.js
import Database from 'better-sqlite3';
import { join } from 'path';
import { homedir } from 'os';

const db = new Database(join(homedir(), '.mytool', 'data.db'));

// Create tables on first run
db.exec(`
  CREATE TABLE IF NOT EXISTS items (
    id INTEGER PRIMARY KEY AUTOINCREMENT,
    name TEXT NOT NULL,
    created_at DATETIME DEFAULT CURRENT_TIMESTAMP
  )
`);

// Fast synchronous queries
const items = db.prepare('SELECT * FROM items').all();

Script to Product Evolution

Growing a script into a real product

When to use: When a personal tool shows promise

Evolution Path

Stage 1: Personal Script

Characteristics:
- Only you use it
- Hardcoded values
- No error handling
- Works on your machine

Time: Hours to days

Stage 2: Shareable Tool

Add:
- README explaining what it does
- Basic error messages
- Config file instead of hardcoding
- Works on similar machines

Time: Days

Stage 3: Public Tool

Add:
- Installation instructions
- Cross-platform support
- Proper error handling
- Version numbers
- Basic tests

Time: Week or two

Stage 4: Product

Add:
- Landing page
- Documentation site
- User support channel
- Analytics (privacy-respecting)
- Payment integration (if monetizing)

Time: Weeks to months

Signs You Should Productize

SignalStrength
Others asking for itStrong
You use it dailyStrong
Solves $100+ problemStrong
Others would payVery strong
Competition exists but sucksStrong
You're embarrassed by itActually good

Sharp Edges

Tool only works in your specific environment

Severity: MEDIUM

Situation: Script fails when you try to share it

Symptoms:

  • Works on my machine
  • Scripts failing for others
  • Path not found errors
  • Command not found errors

Why this breaks: Hardcoded absolute paths. Relies on your installed tools. Assumes your OS/shell. Uses your auth tokens.

Recommended fix:

Making Tools Portable

Common Portability Issues

IssueFix
Hardcoded pathsUse ~ or env vars
Specific shellDeclare shell in shebang
Missing depsCheck and prompt to install
Auth tokensUse config file or env
OS-specificTest on other OS or use cross-platform libs

Path Portability

// Bad
const dataFile = '~/data.json';

// Good
import { homedir } from 'os';
import { join } from 'path';
const dataFile = join(homedir(), '.mytool', 'data.json');

Dependency Checking

import { execSync } from 'child_process';

function checkDep(cmd, installHint) {
  try {
    execSync(`which ${cmd}`, { stdio: 'ignore' });
  } catch {
    console.error(`Missing: ${cmd}`);
    console.error(`Install: ${installHint}`);
    process.exit(1);
  }
}

checkDep('ffmpeg', 'brew install ffmpeg');

Cross-Platform Considerations

import { platform } from 'os';

const isWindows = platform() === 'win32';
const isMac = platform() === 'darwin';
const isLinux = platform() === 'linux';

// Path separator
import { sep } from 'path';
// Use sep instead of hardcoded / or \

Configuration becomes unmanageable

Severity: MEDIUM

Situation: Too many config options making the tool unusable

Symptoms:

  • Config file is huge
  • Users confused by options
  • You forget what options exist
  • Every bug fix adds a flag

Why this breaks: Adding options instead of opinions. Fear of making decisions. Every edge case becomes an option. Config file larger than the tool.

Recommended fix:

Taming Configuration

The Config Hierarchy

Best to worst:
1. Smart defaults (no config needed)
2. Single config file
3. Environment variables
4. Command-line flags
5. Interactive prompts

Use sparingly:
6. Config directory with multiple files
7. Config inheritance/merging

Opinionated Defaults

// Instead of 10 options, pick reasonable defaults
const defaults = {
  outputDir: join(homedir(), '.mytool', 'output'),
  format: 'json',  // Not a flag, just pick one
  maxItems: 100,   // Good enough for most
  verbose: false
};

// Only expose what REALLY needs customization
// "Would I want to change this?" - not "Could someone?"

Config File Pattern

// ~/.mytool/config.json
// Keep it minimal
{
  "apiKey": "xxx",       // Actually needed
  "defaultProject": "main"  // Convenience
}

// Don't do this:
{
  "outputFormat": "json",
  "outputIndent": 2,
  "outputColorize": true,
  "logLevel": "info",
  "logFormat": "pretty",
  "logTimestamp": true,
  // ... 50 more options
}

When to Add Options

Add option if...Don't add if...
Users ask repeatedlyYou imagine someone might want
Security/auth relatedIt's a "nice to have"
Fundamental behavior changeIt's a micro-preference
Environment-specificYou can pick a good default

Personal tool becomes unmaintained

Severity: LOW

Situation: Tool you built is now broken and you don't want to fix it

Symptoms:

  • Script hasn't run in months
  • Don't remember how it works
  • Dependencies outdated
  • Workflow has changed

Why this breaks: Built for old workflow. Dependencies broke. Lost interest. No documentation for yourself.

Recommended fix:

Sustainable Personal Tools

Design for Abandonment

Assume future-you won't remember:
- Why you built this
- How it works
- Where the data is
- What the dependencies do

Build accordingly:
- README with WHY, not just WHAT
- Simple architecture
- Minimal dependencies
- Data in standard formats

Minimal Dependency Strategy

ApproachWhen to Use
Zero depsSimple scripts
Core deps onlyCLI tools
Lock versionsImportant tools
Bundle depsDistribution

Self-Documenting Pattern

#!/usr/bin/env node
/**
 * WHAT: Converts X to Y
 * WHY: Because Z process was manual
 * WHERE: Data in ~/.mytool/
 * DEPS: Needs ffmpeg installed
 *
 * Last used: 2024-01
 * Still works as of: 2024-01
 */

// Tool code here

Graceful Degradation

// When things break, fail helpfully
try {
  await runMainFeature();
} catch (err) {
  console.error('Tool broken. Error:', err.message);
  console.error('');
  console.error('Data location: ~/.mytool/data.json');
  console.error('You can manually access your data there.');
  process.exit(1);
}

When to Let Go

Signs to abandon:
- Haven't used in 6+ months
- Problem no longer exists
- Better tool now exists
- Would rebuild differently

How to abandon gracefully:
- Archive in clear state
- Note why abandoned
- Export data to standard format
- Don't delete (might want later)

Personal tools with security vulnerabilities

Severity: HIGH

Situation: Your personal tool exposes sensitive data or access

Symptoms:

  • API keys in source code
  • Tool accessible on network
  • Credentials in git history
  • Personal data exposed

Why this breaks: "It's just for me" mentality. Credentials in code. No input validation. Accidental exposure.

Recommended fix:

Security in Personal Tools

Common Mistakes

RiskMitigation
API keys in codeUse env vars or config file
Tool exposed on networkBind to localhost only
No input validationValidate even your own input
Logs contain secretsSanitize logging
Git commits with secrets.gitignore config files

Credential Management

// Never in code
const API_KEY = 'sk-xxx'; // BAD

// Environment variable
const API_KEY = process.env.MY_API_KEY;

// Config file (gitignored)
import { readFileSync } from 'fs';
const config = JSON.parse(
  readFileSync(join(homedir(), '.mytool', 'config.json'))
);
const API_KEY = config.apiKey;

Localhost-Only Servers

// If your tool has a web UI
import express from 'express';
const app = express();

// ALWAYS bind to localhost for personal tools
app.listen(3000, '127.0.0.1', () => {
  console.log('Running on http://localhost:3000');
});

// NEVER do this for personal tools:
// app.listen(3000, '0.0.0.0') // Exposes to network!

Before Sharing

Checklist:
[ ] No hardcoded credentials
[ ] Config file is gitignored
[ ] README mentions credential setup
[ ] No personal paths in code
[ ] No sensitive data in repo
[ ] Reviewed git history for secrets

Validation Checks

Hardcoded Absolute Paths

Severity: MEDIUM

Message: Hardcoded absolute path - use homedir() or environment variables.

Fix action: Use os.homedir() or path.join for portable paths

Hardcoded Credentials

Severity: CRITICAL

Message: Potential hardcoded credential - use environment variables or config file.

Fix action: Move to process.env.VAR or external config file (gitignored)

Server Bound to All Interfaces

Severity: HIGH

Message: Server exposed to network - bind to localhost for personal tools.

Fix action: Use '127.0.0.1' or 'localhost' instead of '0.0.0.0'

Missing Error Handling

Severity: MEDIUM

Message: Sync operation without error handling - wrap in try/catch.

Fix action: Add try/catch for graceful error messages

CLI Without Help

Severity: LOW

Message: CLI has no help - future you will forget how to use it.

Fix action: Add.description() and --help to CLI commands

Tool Without README

Severity: LOW

Message: No README - document for your future self.

Fix action: Add README with: what it does, why you built it, how to use it

Debug Console Logs Left In

Severity: LOW

Message: Debug logging left in code - remove or use proper logging.

Fix action: Remove debug logs or use a proper logger with levels

Script Missing Shebang

Severity: LOW

Message: Script missing shebang - won't execute directly.

Fix action: Add #!/usr/bin/env node (or python3) at top of file

Tool Without Version

Severity: LOW

Message: No version tracking - will cause confusion when updating.

Fix action: Add version to package.json and --version flag

Collaboration

Delegation Triggers

  • sell|monetize|SaaS|charge -> micro-saas-launcher (Productizing personal tool)
  • browser extension|chrome extension -> browser-extension-builder (Building browser-based tool)
  • automate|workflow|cron|trigger -> workflow-automation (Automation setup)
  • API|server|database|postgres -> backend (Backend infrastructure)
  • telegram bot -> telegram-bot-builder (Telegram-based tool)
  • AI|GPT|Claude|LLM -> ai-wrapper-product (AI-powered tool)

CLI Tool That Becomes Product

Skills: personal-tool-builder, micro-saas-launcher

Workflow:

1. Build CLI for yourself
2. Share with friends/colleagues
3. Get feedback and iterate
4. Add web UI (optional)
5. Set up payments
6. Launch publicly

Personal Automation Stack

Skills: personal-tool-builder, workflow-automation, backend

Workflow:

1. Identify repetitive task
2. Build script to automate
3. Add triggers (cron, webhook)
4. Store results/logs
5. Monitor and iterate

AI-Powered Personal Tool

Skills: personal-tool-builder, ai-wrapper-product

Workflow:

1. Identify task AI can help with
2. Build minimal wrapper
3. Tune prompts for your use case
4. Add to daily workflow
5. Consider sharing if useful

Browser Tool to Extension

Skills: personal-tool-builder, browser-extension-builder

Workflow:

1. Build bookmarklet or userscript
2. Validate it solves the problem
3. Convert to proper extension
4. Add to Chrome/Firefox store
5. Share with others

Related Skills

Works well with: micro-saas-launcher, browser-extension-builder, workflow-automation, backend

When to Use

  • User mentions or implies: build a tool
  • User mentions or implies: personal tool
  • User mentions or implies: scratch my itch
  • User mentions or implies: solve my problem
  • User mentions or implies: CLI tool
  • User mentions or implies: local app
  • User mentions or implies: automate my
  • User mentions or implies: build for myself

Limitations

  • Use this skill only when the task clearly matches the scope described above.
  • Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
  • Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

平台分布

Claude Code

29.02%
按下载量换算1,042

OpenCode

26.44%
按下载量换算949

Gemini CLI

17.03%
按下载量换算611

Antigravity

12.78%
按下载量换算459

Cursor

8.65%
按下载量换算311

Codex

3.93%
按下载量换算141

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。

继续浏览同类 Skills