Token导航 LogoToken导航TokenDH.com
研究检索执行命令github未标认证来源可访问clear审计未展示

skill-builder技能构建

Agent Skill

skill-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。

总安装

269

周安装

11

GitHub Stars

公开资料未说明

下载量

86
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

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:skill-builder(技能构建)
来源仓库:https://github.com/dvorkinguy/claude-skills-agents
仓库路径:skills/skill-builder
安装命令:
npx skills add dvorkinguy/claude-skills-agents --skill "skill-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。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 dvorkinguy/claude-skills-agents --skill "skill-builder"

简介

skill-builder 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。

  • 适用于技能开发、工具集成或代理能力扩展场景,支持基于关键词的任务匹配。
  • 通过 npx skills add 命令从 GitHub 仓库安装,便于本地或云端部署使用。
  • 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。
  • 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

name
skill-builder
description
Guide for creating effective Claude Code skills. Use when users want to create a new skill, update an existing skill, package a skill for distribution, or need guidance on skill structure and best practices. Triggers include "create skill", "build skill", "new skill", "make skill", "skill template", "package skill", or working with SKILL.md files.

Skill Builder

Create Claude Code skills that extend capabilities with specialized knowledge, workflows, and tools.

Quick Commands

TaskCommand
New skillpython scripts/init_skill.py <name> --path <dir>
Validatepython scripts/quick_validate.py <skill-dir>
Packagepython scripts/package_skill.py <skill-dir> [output-dir]

Core Principles

  1. Concise is key - Claude is smart; only add what it doesn't know
  2. Progressive disclosure - Metadata always loaded, SKILL.md on trigger, references on-demand
  3. Appropriate freedom - High for flexible tasks, low for fragile operations

Skill Creation Process

Step 1: Clarify Requirements

Ask the user:

  • What task should the skill automate?
  • What triggers should invoke it? (keywords, file types, scenarios)
  • What outputs are expected?
  • Can you show example use cases?

Step 2: Plan Resources

Analyze each use case to identify:

  • scripts/ - Repetitive code that should be deterministic
  • references/ - Documentation too long for SKILL.md
  • assets/ - Templates, images, boilerplate for output

Step 3: Initialize

python scripts/init_skill.py my-skill --path ~/.claude/skills

Step 4: Edit SKILL.md

Frontmatter (Critical)

---
name: my-skill
description: What this skill does and WHEN to use it. Include triggers, file types, scenarios. This is the primary trigger mechanism. Max 1024 chars.
---

Description rules:

  • Include BOTH what the skill does AND when to use it
  • List specific triggers (file types, keywords, scenarios)
  • No < or > characters
  • Max 1024 characters

Body Patterns

Pattern 1: Workflow-Based (sequential processes)

## Workflow
1. Analyze input -> Run `scripts/analyze.py`
2. Transform -> Apply rules
3. Output -> Save result

Pattern 2: Task-Based (tool collections)

## Tasks
### Create New
[instructions]
### Edit Existing
[instructions]

Pattern 3: Reference-Based (standards/guidelines)

## Guidelines
### Colors
### Typography

Step 5: Add Resources

Scripts (scripts/)

For deterministic, repetitive operations.

  • Test scripts before packaging
  • Make executable: chmod +x script.py

References (references/)

Documentation loaded on-demand.

  • For content >100 lines, add table of contents
  • Split by domain/variant when >500 lines
  • Avoid duplication with SKILL.md

Assets (assets/)

Files used in output (not loaded into context).

  • Templates, images, fonts
  • Boilerplate code

Step 6: Validate & Package

python scripts/quick_validate.py my-skill/
python scripts/package_skill.py my-skill/ ./dist

Frontmatter Rules

FieldRequiredLimitNotes
nameYes64 charshyphen-case only (a-z, 0-9, -)
descriptionYes1024 charsNo < or > characters
allowed-toolsNo-Restrict tool access

Progressive Disclosure Patterns

Keep SKILL.md under 500 lines. Split to references/ when needed.

High-level guide with references:

## Quick Start
[Essential workflow]

## Advanced
- **Forms**: See `references/forms.md`
- **API**: See `references/api.md`

Domain-specific organization:

skill/
├── SKILL.md (overview + navigation)
└── references/
    ├── aws.md
    ├── gcp.md
    └── azure.md

What NOT to Include

  • README.md (redundant with SKILL.md)
  • CHANGELOG.md
  • INSTALLATION_GUIDE.md
  • User-facing documentation

Skills are for AI agents, not humans.

References

  • references/skill-creation-guide.md - Complete skill creation guide
  • references/output-patterns.md - Template and example patterns
  • references/workflows.md - Sequential and conditional workflow patterns

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

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

31.39%
按下载量换算27

Antigravity

21.9%
按下载量换算19

Gemini CLI

18.4%
按下载量换算16

windsurf

12.24%
按下载量换算11

OpenCode

8.23%
按下载量换算7

Codex

3.6%
按下载量换算3

安全审计

暂无安全审计结果可展示。

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 dvorkinguy/claude-skills-agents --skill "skill-builder" 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