Token导航 LogoToken导航

Agent Skill · 研究检索

local-skills-setup(本地技能设置)

local-skills-setup 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。

local-skills-setup 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,权限为需要联网;许可证为 MIT,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 51,060。

总安装 594周安装 25GitHub Stars 31,988下载 208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:local-skills-setup(本地技能设置)
来源仓库:https://github.com/yeachan-heo/oh-my-claudecode
仓库路径:skills/local-skills-setup
安装命令:
npx skills add https://github.com/yeachan-heo/oh-my-claudecode --skill local-skills-setup
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/yeachan-heo/oh-my-claudecode --skill local-skills-setup

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。

local-skills-setup 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。适用于需要根据关键词或任务场景从本地资源中获取信息的场景。通过安装命令添加到宿主环境后,可结合原始 README 核验具体用法。

安装前需确认权限范围、维护状态,避免触发联网或文件读写操作。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。

来源仓库内路径:skills/local-skills-setup

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/yeachan-heo/oh-my-claudecode --skill local-ski…;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

    它需要需要联网,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

local-skills-setup 怎么安装?

在终端执行 npx skills add https://github.com/yeachan-heo/oh-my-claudecode --skill local-skills-setup。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

local-skills-setup 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。

local-skills-setup 需要什么权限?

权限级别是需要联网。该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

local-skills-setup 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。

local-skills-setup 用的人多吗?

公开数据是累计安装 594 次,近一周安装 25 次,下载 208 次,GitHub 31,988 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。

local-skills-setup 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。
  • 来源仓库有 31,988 星,维护活跃度相对可查。

先想清楚的地方

  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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

  • 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
local-skills-setup

本地技能设置 · 本页

59431,988需要联网MIT有提醒

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

Local Skills Setup

This skill provides a guided wizard for setting up and managing your local learned skills. Skills are reusable problem-solving patterns that Claude automatically applies when it detects matching triggers.

Why Local Skills?

Local skills allow you to capture hard-won insights and solutions that are specific to your codebase or workflow:

  • Project-level skills (.omc/skills/) - Version-controlled with your repo
  • User-level skills (~/.claude/skills/omc-learned/) - Portable across all your projects

When you solve a tricky bug or discover a non-obvious workaround, you can extract it as a skill. Claude will automatically detect and apply these skills in future conversations when it sees matching triggers.

Interactive Workflow

Step 1: Directory Check and Setup

First, check if skill directories exist and create them if needed:

# Check and create user-level skills directory
USER_SKILLS_DIR="$HOME/.claude/skills/omc-learned"
if [ -d "$USER_SKILLS_DIR" ]; then
  echo "User skills directory exists: $USER_SKILLS_DIR"
else
  mkdir -p "$USER_SKILLS_DIR"
  echo "Created user skills directory: $USER_SKILLS_DIR"
fi

# Check and create project-level skills directory
PROJECT_SKILLS_DIR=".omc/skills"
if [ -d "$PROJECT_SKILLS_DIR" ]; then
  echo "Project skills directory exists: $PROJECT_SKILLS_DIR"
else
  mkdir -p "$PROJECT_SKILLS_DIR"
  echo "Created project skills directory: $PROJECT_SKILLS_DIR"
fi

Step 2: Skill Scan and Inventory

Scan both directories and show a comprehensive inventory:

# Scan user-level skills
echo "=== USER-LEVEL SKILLS (~/.claude/skills/omc-learned/) ==="
if [ -d "$HOME/.claude/skills/omc-learned" ]; then
  USER_COUNT=$(find "$HOME/.claude/skills/omc-learned" -name "*.md" 2>/dev/null | wc -l)
  echo "Total skills: $USER_COUNT"

  if [ $USER_COUNT -gt 0 ]; then
    echo ""
    echo "Skills found:"
    find "$HOME/.claude/skills/omc-learned" -name "*.md" -type f -exec sh -c '
      FILE="$1"
      NAME=$(grep -m1 "^name:" "$FILE" 2>/dev/null | sed "s/name: //")
      DESC=$(grep -m1 "^description:" "$FILE" 2>/dev/null | sed "s/description: //")
      MODIFIED=$(stat -c "%y" "$FILE" 2>/dev/null || stat -f "%Sm" "$FILE" 2>/dev/null)
      echo "  - $NAME"
      [ -n "$DESC" ] && echo "    Description: $DESC"
      echo "    Modified: $MODIFIED"
      echo ""
    ' sh {} \;
  fi
else
  echo "Directory not found"
fi

echo ""
echo "=== PROJECT-LEVEL SKILLS (.omc/skills/) ==="
if [ -d ".omc/skills" ]; then
  PROJECT_COUNT=$(find ".omc/skills" -name "*.md" 2>/dev/null | wc -l)
  echo "Total skills: $PROJECT_COUNT"

  if [ $PROJECT_COUNT -gt 0 ]; then
    echo ""
    echo "Skills found:"
    find ".omc/skills" -name "*.md" -type f -exec sh -c '
      FILE="$1"
      NAME=$(grep -m1 "^name:" "$FILE" 2>/dev/null | sed "s/name: //")
      DESC=$(grep -m1 "^description:" "$FILE" 2>/dev/null | sed "s/description: //")
      MODIFIED=$(stat -c "%y" "$FILE" 2>/dev/null || stat -f "%Sm" "$FILE" 2>/dev/null)
      echo "  - $NAME"
      [ -n "$DESC" ] && echo "    Description: $DESC"
      echo "    Modified: $MODIFIED"
      echo ""
    ' sh {} \;
  fi
else
  echo "Directory not found"
fi

# Summary
TOTAL=$((USER_COUNT + PROJECT_COUNT))
echo "=== SUMMARY ==="
echo "Total skills across all directories: $TOTAL"

Step 3: Quick Actions Menu

After scanning, use the AskUserQuestion tool to offer these options:

Question: "What would you like to do with your local skills?"

Options:

  1. Add new skill - Start the skill creation wizard
  2. List all skills with details - Show comprehensive skill inventory with triggers
  3. Scan conversation for patterns - Analyze current conversation for skill-worthy patterns
  4. Import skill - Import a skill from URL or paste content
  5. Done - Exit the wizard
Option 1: Add New Skill

If user chooses "Add new skill", invoke the learner skill:

Use the Skill tool to invoke: learner

This will guide them through the extraction process with quality validation.

Option 2: List All Skills with Details

Show detailed information including trigger keywords:

echo "=== DETAILED SKILL INVENTORY ==="
echo ""

# Function to show skill details
show_skill_details() {
  FILE="$1"
  LOCATION="$2"

  echo "---"
  echo "Location: $LOCATION"
  echo "File: $(basename "$FILE")"

  # Extract frontmatter
  NAME=$(grep -m1 "^name:" "$FILE" 2>/dev/null | sed "s/name: //")
  DESC=$(grep -m1 "^description:" "$FILE" 2>/dev/null | sed "s/description: //")
  TRIGGERS=$(grep -m1 "^triggers:" "$FILE" 2>/dev/null | sed "s/triggers: //")
  QUALITY=$(grep -m1 "^quality:" "$FILE" 2>/dev/null | sed "s/quality: //")

  [ -n "$NAME" ] && echo "Name: $NAME"
  [ -n "$DESC" ] && echo "Description: $DESC"
  [ -n "$TRIGGERS" ] && echo "Triggers: $TRIGGERS"
  [ -n "$QUALITY" ] && echo "Quality: $QUALITY"

  # Last modified
  MODIFIED=$(stat -c "%y" "$FILE" 2>/dev/null | cut -d. -f1 || stat -f "%Sm" "$FILE" 2>/dev/null)
  echo "Last modified: $MODIFIED"
  echo ""
}

# Export function for subshell
export -f show_skill_details

# Show user-level skills
if [ -d "$HOME/.claude/skills/omc-learned" ]; then
  echo "USER-LEVEL SKILLS:"
  find "$HOME/.claude/skills/omc-learned" -name "*.md" -type f -exec bash -c 'show_skill_details "$0" "user-level"' {} \;
fi

# Show project-level skills
if [ -d ".omc/skills" ]; then
  echo "PROJECT-LEVEL SKILLS:"
  find ".omc/skills" -name "*.md" -type f -exec bash -c 'show_skill_details "$0" "project-level"' {} \;
fi
Option 3: Scan Conversation for Patterns

Analyze the current conversation context to identify potential skill-worthy patterns. Look for:

  • Recent debugging sessions with non-obvious solutions
  • Tricky bugs that required investigation
  • Codebase-specific workarounds discovered
  • Error patterns that took time to resolve

Report findings and ask if user wants to extract any as skills.

Option 4: Import Skill

Ask user to provide either:

  • URL: Download skill from a URL (e.g., GitHub gist)
  • Paste content: Paste skill markdown content directly

Then ask for scope:

  • User-level (~/.claude/skills/omc-learned/) - Available across all projects
  • Project-level (.omc/skills/) - Only for this project

Validate the skill format and save to the chosen location.

Step 4: Skill Templates

Provide quick templates for common skill types. When user wants to create a skill, offer these starting points:

Error Solution Template
---
id: error-[unique-id]
name: [Error Name]
description: Solution for [specific error in specific context]
source: conversation
triggers: ["error message fragment", "file path", "symptom"]
quality: high
---

# [Error Name]

## The Insight
What is the underlying cause of this error? What principle did you discover?

## Why This Matters
What goes wrong if you don't know this? What symptom led here?

## Recognition Pattern
How do you know when this applies? What are the signs?
- Error message: "[exact error]"
- File: [specific file path]
- Context: [when does this occur]

## The Approach
Step-by-step solution:
1. [Specific action with file/line reference]
2. [Specific action with file/line reference]
3. [Verification step]

## Example
\`\`\`typescript
// Before (broken)
[problematic code]

// After (fixed)
[corrected code]
\`\`\`
Workflow Skill Template
---
id: workflow-[unique-id]
name: [Workflow Name]
description: Process for [specific task in this codebase]
source: conversation
triggers: ["task description", "file pattern", "goal keyword"]
quality: high
---

# [Workflow Name]

## The Insight
What makes this workflow different from the obvious approach?

## Why This Matters
What fails if you don't follow this process?

## Recognition Pattern
When should you use this workflow?
- Task type: [specific task]
- Files involved: [specific patterns]
- Indicators: [how to recognize]

## The Approach
1. [Step with specific commands/files]
2. [Step with specific commands/files]
3. [Verification]

## Gotchas
- [Common mistake and how to avoid it]
- [Edge case and how to handle it]
Code Pattern Template
---
id: pattern-[unique-id]
name: [Pattern Name]
description: Pattern for [specific use case in this codebase]
source: conversation
triggers: ["code pattern", "file type", "problem domain"]
quality: high
---

# [Pattern Name]

## The Insight
What's the key principle behind this pattern?

## Why This Matters
What problems does this pattern solve in THIS codebase?

## Recognition Pattern
When do you apply this pattern?
- File types: [specific files]
- Problem: [specific problem]
- Context: [codebase-specific context]

## The Approach
Decision-making heuristic, not just code:
1. [Principle-based step]
2. [Principle-based step]

## Example
\`\`\`typescript
[Illustrative example showing the principle]
\`\`\`

## Anti-Pattern
What NOT to do and why:
\`\`\`typescript
[Common mistake to avoid]
\`\`\`
Integration Skill Template
---
id: integration-[unique-id]
name: [Integration Name]
description: How [system A] integrates with [system B] in this codebase
source: conversation
triggers: ["system name", "integration point", "config file"]
quality: high
---

# [Integration Name]

## The Insight
What's non-obvious about how these systems connect?

## Why This Matters
What breaks if you don't understand this integration?

## Recognition Pattern
When are you working with this integration?
- Files: [specific integration files]
- Config: [specific config locations]
- Symptoms: [what indicates integration issues]

## The Approach
How to work with this integration correctly:
1. [Configuration step with file paths]
2. [Setup step with specific details]
3. [Verification step]

## Gotchas
- [Integration-specific pitfall #1]
- [Integration-specific pitfall #2]

Usage Modes

Direct Command Mode

When invoked with an argument, skip the interactive wizard:

  • /oh-my-claudecode:local-skills-setup list - Show detailed skill inventory
  • /oh-my-claudecode:local-skills-setup add - Start skill creation (invoke learner)
  • /oh-my-claudecode:local-skills-setup scan - Scan both skill directories

Interactive Mode

When invoked without arguments, run the full guided wizard (Steps 1-4).

Skill Quality Guidelines

Remind users that good skills are:

  1. Non-Googleable - Can't easily find via search

- BAD: "How to read files in TypeScript" ❌ - GOOD: "This codebase uses custom path resolution requiring fileURLToPath" ✓

  1. Context-Specific - References actual files/errors from THIS codebase

- BAD: "Use try/catch for error handling" ❌ - GOOD: "The aiohttp proxy in server.py:42 crashes on ClientDisconnectedError" ✓

  1. Actionable with Precision - Tells exactly WHAT to do and WHERE

- BAD: "Handle edge cases" ❌ - GOOD: "When seeing 'Cannot find module' in dist/, check tsconfig.json moduleResolution" ✓

  1. Hard-Won - Required significant debugging effort

- BAD: Generic programming patterns ❌ - GOOD: "Race condition in worker.ts - Promise.all at line 89 needs await" ✓

Benefits of Local Skills

When introducing the skill system, explain these benefits:

Automatic Application: Claude detects triggers and applies skills automatically - no need to remember or search for solutions.

Version Control: Project-level skills (.omc/skills/) are committed with your code, so the whole team benefits.

Evolving Knowledge: Skills improve over time as you discover better approaches and refine triggers.

Reduced Token Usage: Instead of re-solving the same problems, Claude applies known patterns efficiently.

Codebase Memory: Preserves institutional knowledge that would otherwise be lost in conversation history.

Related Skills

  • /oh-my-claudecode:learner - Extract a skill from current conversation
  • /oh-my-claudecode:note - Save quick notes (less formal than skills)
  • /oh-my-claudecode:deepinit - Generate AGENTS.md codebase hierarchy

Example Session

Show users what a typical session looks like:

> /oh-my-claudecode:local-skills-setup

Checking skill directories...
✓ User skills directory exists: ~/.claude/skills/omc-learned/
✓ Project skills directory exists: .omc/skills/

Scanning for skills...

=== USER-LEVEL SKILLS ===
Total skills: 3
  - async-network-error-handling
    Description: Pattern for handling independent I/O failures in async network code
    Modified: 2026-01-20 14:32:15

  - esm-path-resolution
    Description: Custom path resolution in ESM requiring fileURLToPath
    Modified: 2026-01-19 09:15:42

=== PROJECT-LEVEL SKILLS ===
Total skills: 5
  - session-timeout-fix
    Description: Fix for sessionId undefined after restart in session.ts
    Modified: 2026-01-22 16:45:23

  - build-cache-invalidation
    Description: When to clear TypeScript build cache to fix phantom errors
    Modified: 2026-01-21 11:28:37

=== SUMMARY ===
Total skills: 8

What would you like to do?
1. Add new skill
2. List all skills with details
3. Scan conversation for patterns
4. Import skill
5. Done

Tips for Users

  • Run /oh-my-claudecode:local-skills-setup scan periodically to review your skill library
  • After solving a tricky bug, immediately run learner to capture it
  • Use project-level skills for codebase-specific knowledge
  • Use user-level skills for general patterns that apply everywhere
  • Review and refine triggers over time to improve matching accuracy

平台分布

Claude Code

26.49%
按下载量换算55

OpenCode

25.51%
按下载量换算53

Gemini CLI

19.2%
按下载量换算40

Antigravity

12.8%
按下载量换算27

Cursor

6.84%
按下载量换算14

Codex

3.32%
按下载量换算7

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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。来源安全扫描存在提醒或未通过项,安装前请自行复核。

继续浏览同类 Skills