Token导航 LogoToken导航

Agent Skill · 研究检索

skill-maker(技能制造者)

skill-maker 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。

skill-maker 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,权限为接触敏感数据;许可证为 MIT,暂无第三方扫描结果,按安装量在研究检索类的 73,675 个 Skill 里排第 72,323。

总安装 188周安装 8GitHub Stars 公开资料未说明下载 66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:skill-maker(技能制造者)
来源仓库:https://github.com/jwynia/agent-skills
仓库路径:skills/skill-maker
安装命令:
npx skills add jwynia/agent-skills --skill "skill-maker"
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。

AgentSkills.tonpx skills
npx skills add jwynia/agent-skills --skill "skill-maker"

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。

用于发现和安装 AI 代理技能的工具集合。适合扩展 Agent 能力范围和自动化任务处理。通过 GitHub 安装,支持主流宿主环境集成。建议在安装前评估技能的安全性和维护状态。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-maker 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/skill-maker

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 jwynia/agent-skills --skill "skill-maker";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

    它需要接触敏感数据,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

skill-maker 怎么安装?

在终端执行 npx skills add jwynia/agent-skills --skill "skill-maker"。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

skill-maker 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。

skill-maker 需要什么权限?

权限级别是接触敏感数据。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。

skill-maker 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。

skill-maker 用的人多吗?

公开数据是累计安装 188 次,近一周安装 8 次,下载 66 次。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。

skill-maker 安全吗?

没有查到第三方扫描结果,安全性无法从公开资料判断,安装前请自行审查源码、依赖和命令行为。

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。
  • 许可证明确为 MIT,商用和分发边界好判断。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 没有第三方扫描结果,安全性只能靠自己读源码判断。
  • 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。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 要求有安全扫描结论才肯安装的团队。
  • 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
skill-maker

技能制造者 · 本页

188—接触敏感数据MIT未收录

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

Skill Maker

Create skills that follow the agentskills.io specification. This skill provides a structured workflow for gathering requirements, scaffolding the skill structure, and validating the result.

A skill is a directory containing a SKILL.md file with YAML frontmatter and Markdown instructions. Skills extend agent capabilities with specialized knowledge, workflows, and bundled resources.

Skill Creation Workflow

Follow these four phases in order:

Phase 1: Requirements Gathering

Before writing any code, understand what the skill needs to do. Use the interactive questions in the next section to gather:

  • The task or workflow the skill enables
  • Concrete usage examples (what users will say)
  • Required resources (scripts, references, assets)
  • Trigger phrases for the description

Phase 2: Planning

With requirements gathered, plan the skill structure:

  1. Choose a name (lowercase, hyphens, max 64 chars, matches folder)
  2. Draft the description (what + when + keywords, max 1024 chars)
  3. Decide which optional directories are needed:

- scripts/ - Executable code for repetitive/deterministic tasks - references/ - Documentation loaded as needed - assets/ - Templates, images, data files for output

  1. Outline the SKILL.md body sections

Use assets/planning-template.md for structured planning.

Phase 3: Scaffolding

Create the skill structure using the scaffold script:

deno run --allow-write --allow-read scripts/scaffold.ts \
  --name "my-skill" \
  --description "What this skill does and when to use it." \
  --path "./skills/domain/" \
  --with-scripts \
  --with-references

This creates:

my-skill/
├── SKILL.md           # Starter template with frontmatter
├── scripts/           # If --with-scripts
└── references/        # If --with-references

Then edit the generated SKILL.md to add instructions.

Phase 4: Validation

Validate the completed skill:

deno run --allow-read --allow-run scripts/validate-skill.ts ./path/to/skill

The validator checks:

  • Frontmatter validity (name, description constraints)
  • Name matches directory name
  • Line count (warns if >500 lines)
  • Referenced files exist
  • Description includes trigger phrases

Fix any issues and re-validate.

Quick Start

To create a minimal skill immediately:

  1. Create the directory: mkdir -p skills/domain/my-skill
  2. Copy the template: cp assets/skill-md-template.md skills/domain/my-skill/SKILL.md
  3. Edit the SKILL.md frontmatter and body
  4. Validate: deno run --allow-read --allow-run scripts/validate-skill.ts./skills/domain/my-skill

Interactive Requirements Gathering

When a user requests a new skill, ask these questions:

Understanding the Task

  • "What task or workflow should this skill enable?"
  • "Can you give 2-3 concrete examples of how this skill would be used?"
  • "What would a user say that should trigger this skill?"

Identifying Resources

  • "Will this skill need executable scripts for repetitive tasks?"
  • "Is there reference documentation the skill should include?"
  • "Are there templates or assets the skill needs to produce output?"

Defining Scope

  • "What should this skill NOT handle? (Out of scope)"
  • "Are there related skills this should connect to?"

Drafting the Description

Based on gathered information, draft a description following this pattern:

"[What the skill does]. This skill should be used when the user asks to [trigger phrase 1], [trigger phrase 2], [trigger phrase 3], or [scenario]. Keywords: [relevant terms]."

Key Principles

Progressive Disclosure

Structure skills for efficient context usage:

LevelContentSize
Metadataname + description~100 tokens (always loaded)
InstructionsSKILL.md body<5000 tokens (loaded on activation)
Resourcesscripts/, references/, assets/As needed

Keep SKILL.md under 500 lines. Move detailed content to references/.

Strong Trigger Descriptions

Write descriptions in third person with specific phrases:

# Good
description: "This skill should be used when the user asks to 'create a PDF',
  'extract text from PDF', 'merge PDFs', or mentions PDF processing."

# Bad
description: "Helps with PDFs."

Imperative Writing Style

Write instructions in imperative form, not second person:

# Good
Run the validation script before committing changes.
Configure the API key in the environment.

# Bad
You should run the validation script.
You need to configure the API key.

Scripts Reference

scaffold.ts

Creates skill folder structure and starter files.

deno run --allow-write --allow-read scripts/scaffold.ts \
  --name "skill-name" \
  --description "Description text" \
  --path "./target/directory/" \
  [--with-scripts] [--with-references] [--with-assets]

Options:

  • --name (required): Skill name (validated for format)
  • --description (required): Initial description
  • --path (required): Parent directory for the skill
  • --with-scripts: Create scripts/ directory
  • --with-references: Create references/ directory
  • --with-assets: Create assets/ directory

validate-skill.ts

Validates a skill against the agentskills.io specification.

deno run --allow-read --allow-run scripts/validate-skill.ts ./path/to/skill

Performs checks beyond the basic spec:

  • Calls skills-ref validate if available
  • Warns on large files (>500 lines)
  • Checks for referenced file existence
  • Detects second-person language patterns

Additional Resources

Reference Files

For detailed guidance, consult:

  • references/spec-quick-reference.md - Condensed agentskills.io specification
  • references/progressive-disclosure.md - Content organization strategies
  • references/examples-gallery.md - Annotated examples of well-structured skills

Asset Files

Templates for skill creation:

  • assets/skill-md-template.md - Complete SKILL.md starter template
  • assets/planning-template.md - Requirements gathering worksheet

External Resources

平台分布

kilo

28.77%
按下载量换算19

windsurf

23.59%
按下载量换算16

amp

18.38%
按下载量换算12

clawdbot

11.94%
按下载量换算8

OpenCode

7.44%
按下载量换算5

Cursor

3.64%
按下载量换算2

安全审计

暂无安全审计结果可展示。

权限和风险

接触敏感数据

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills