Token导航 LogoToken导航

Agent Skill · 研究检索

explainer(explainer 搜索)

explainer 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。

explainer 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,权限为只读;许可证为 MIT,第三方扫描有未通过项,按安装量在研究检索类的 73,675 个 Skill 里排第 64,698。

总安装 294周安装 12GitHub Stars 142下载 94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:explainer(explainer 搜索)
来源仓库:https://github.com/thebushidocollective/han
仓库路径:skills/explainer
安装命令:
npx skills add https://github.com/thebushidocollective/han --skill explainer
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/thebushidocollective/han --skill explainer

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。

explainer 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据关键词、任务场景或来源线索快速定位候选结果。通过 npx skills add 命令从指定 GitHub 路径安装并使用。安装前需确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

建议结合来源仓库和原始 README 核验具体用法和功能边界。

来源仓库内路径:skills/explainer

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/thebushidocollective/han --skill explainer;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

explainer 怎么安装?

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

explainer 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。

explainer 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

explainer 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。

explainer 用的人多吗?

公开数据是累计安装 294 次,近一周安装 12 次,下载 94 次,GitHub 142 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。

explainer 安全吗?

第三方扫描有未通过项,风险较高。除非你确认过源码,否则不建议直接安装到常用环境。

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。
  • 标注为只读权限,试用阶段不容易误改本地文件。

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。
  • 只想先试试效果、不希望它改动本地文件的人。

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
explainer

explainer 搜索 · 本页

294142只读MIT未通过

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

Explainer Skill

Create clear, insightful explanations of code, concepts, and technical decisions.

Core Principle

Clarity over completeness. An explanation is only useful if it's understood.

Explanation Process

1. Understand the Question

What are they really asking?

  • Literal question: "What does this do?"
  • Deeper question: "Why does this exist?" or "How does this work?"
  • Context matters: Junior dev vs senior dev needs different explanations

2. Gather Full Context

Read before explaining:

  • The code in question (obvious)
  • Surrounding code (function, class, module)
  • Related code (what calls it, what it calls)
  • Tests (reveal intended behavior)
  • Comments (capture original reasoning)
  • Git history (understand evolution)

Anti-pattern: Explaining code you only partially understand

3. Structure the Explanation

Start broad, then narrow:

  1. One-sentence summary: What it does in plain English
  2. Purpose: Why it exists, what problem it solves
  3. High-level approach: How it solves the problem (conceptually)
  4. Implementation details: Specific code patterns, algorithms
  5. Edge cases: Unusual scenarios handled
  6. Related concepts: Connections to other parts of the system

4. Choose the Right Level

Match explanation to context:

For "What does this do?"

BAD: "It's a reducer function that takes state and action..."
GOOD: "This manages the shopping cart state. When you add/remove
       items, it updates the cart and recalculates the total."

For "How does this work?"

BAD: "It just loops through items and sums prices."
GOOD: "It iterates through cart items, applies discounts to each,
       then sums the discounted prices. Tax is calculated on the
       subtotal, not individual items, to avoid rounding errors."

For "Why this approach?"

BAD: "Because it's faster."
GOOD: "We chose this over X because: (1) handles async updates
       correctly, (2) prevents race conditions when multiple users
       modify the cart, (3) makes testing easier. Trade-off is
       slightly more complex code."

Explanation Patterns

Code Explanation Format

## What it does

[One-sentence plain English summary]

## Purpose

[Why this code exists, what problem it solves]

## How it works

1. [High-level step 1]
2. [High-level step 2]
3. [High-level step 3]

### Key details

- [Important implementation detail 1]
- [Important implementation detail 2]

### Example

[Concrete example showing usage]

## Related

- [Link to related code/docs]

Concept Explanation Format

## [Concept Name]

[One-sentence definition]

### Why it matters

[Practical importance, when you'd use it]

### How it works

[High-level explanation without jargon]

### Example

[Concrete, relatable example]

### In our codebase

[Where we use this concept, with links]

### Common pitfalls

[What to watch out for]

Decision Explanation Format

## Decision: [What was decided]

### Context

[Situation that required a decision]

### Options considered

1. **[Option 1]**: [Brief description]
   - Pros: [Key benefits]
   - Cons: [Key drawbacks]

2. **[Option 2]**: [Brief description]
   - Pros: [Key benefits]
   - Cons: [Key drawbacks]

### Choice: [Selected option]

**Rationale:**
[Why this option was chosen over others]

### Trade-offs accepted

[What we gave up by choosing this approach]

### References

[Links to discussions, docs, related decisions]

Writing Guidelines

Be Concrete

Bad: "This uses a common design pattern for state management" Good: "This uses the Redux pattern: all state changes go through a central store via actions and reducers"

Use Analogies (When Helpful)

Example: "Think of this like a restaurant kitchen:

  • Actions are customer orders
  • Reducers are chefs preparing food
  • Store is the completed order waiting area"

Warning: Don't force analogies. If it's clearer without, skip it.

Show, Don't Just Tell

// BAD explanation:
// "This function filters and maps the array"

// GOOD explanation:
// "This function finds all active users and formats them for display:
const displayUsers = users
  .filter(u => u.status === 'active')  // Only show active users
  .map(u => ({                         // Convert to display format
    id: u.id,
    name: `${u.firstName} ${u.lastName}`,
    joined: formatDate(u.createdAt)
  }))

Avoid Jargon (Unless Explaining Jargon)

Bad: "This leverages a memoized selector with referential equality optimization" Good: "This caches the filtered list so we don't recalculate it every render, improving performance"

Exception: When explaining what jargon means: "Memoization means caching function results so we don't recompute the same thing twice."

Common Explanation Scenarios

Explaining Error Messages

## Error: "Cannot read property 'map' of undefined"

**What happened:**
You're trying to use `.map()` on something that's `undefined`.

**Where:** [Link to line]

**Why:**
The `users` array hasn't loaded yet (async), so it's `undefined`
when the component first renders.

**Fix:**

// Before users.map(u => ...)

// After {users?.map(u => ...) || <Loading />}


**Why this works:** `?.` is optional chaining - it only calls `.map()` if `users` exists. If not, it shows a loading state instead.

Explaining "Why Not X?"

## Why not use X?

**Context:** [What X is]

**Reasons we chose Y instead:**

1. **[Primary reason]**: [Explanation]
2. **[Secondary reason]**: [Explanation]
3. **[Tertiary reason]**: [Explanation]

**Trade-offs:**
Y is slower than X, but the reliability benefit outweighs the
50ms performance difference.

**When X would be better:**
If we needed real-time updates (< 100ms), X would be the right choice.

Explaining Legacy Code

## Legacy: [Module Name]

**What it does:** [Current function]

**History:**
This was written in 2019 when [context]. At the time, [justification].

**Why it looks odd today:**
Modern approaches would use [better pattern], but this works and
changing it isn't worth the risk.

**If you must modify it:**
1. [Key constraint to preserve]
2. [Another constraint]
3. Add tests first (none exist currently)

**Related:** See [modern equivalent] for new code.

Anti-Patterns

❌ Assuming Knowledge

BAD: "This uses HOCs to inject props via connect()"
GOOD: "This connects the component to Redux, giving it access
       to the store data it needs. See: [Redux docs link]"

❌ Over-explaining Obvious Code

// Don't explain this:
const total = price * quantity  // Multiplies price by quantity

// Do explain this:
const total = price * quantity * (1 - discount) * TAX_RATE
// Discount applied before tax, per accounting requirements

❌ Vague References

BAD: "Similar to what we do in other places"
GOOD: "Similar to UserService.findActive() in services/user.ts:45"

❌ Explaining *How* When Asked *Why*

Question: "Why do we use a Set here instead of an Array?"

Bad: "A Set is initialized with new Set() and supports .add() and .has() methods..."

Good: "A Set automatically removes duplicates. We need unique user IDs, and Set handles that for us. An Array would require manual duplicate checking."

Verification

After explaining, check:

  • ✅ Does this answer the actual question asked?
  • ✅ Is it understandable without reading it 3 times?
  • ✅ Are technical terms explained or linked?
  • ✅ Are there concrete examples?
  • ✅ Does it avoid unnecessary complexity?
  • ✅ Would someone new to the codebase understand this?

Integration with Other Skills

  • Use code-reviewer skill when explaining code quality issues
  • Use proof-of-work skill when explaining behavior (show actual output)
  • Use simplicity-principles skill when explaining why simpler is better
  • Reference other skills when explaining patterns they cover

Remember

The best explanation is the one that's understood, not the one that's most complete.

If in doubt, start simple and add detail only when needed.

平台分布

Claude Code

26.58%
按下载量换算25

OpenCode

21.81%
按下载量换算21

Codex

16.59%
按下载量换算16

Antigravity

13.58%
按下载量换算13

windsurf

7.71%
按下载量换算7

Gemini CLI

2.95%
按下载量换算3

安全审计

Gen Agent Trust Hub

未通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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