learn

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

learn

Research any topic by gathering online resources and creating a comprehensive learning guide with RAG-optimized indexes.

通过收集线上资源调研任意主题，使用RAG优化的索引创建全面的学习指南。

Parse Arguments

参数解析

javascript

const args = '$ARGUMENTS'.split(' ').filter(Boolean);
const depth = args.find(a => a.startsWith('--depth='))?.split('=')[1] || 'medium';
const topic = args.filter(a => !a.startsWith('--')).join(' ');

javascript

const args = '$ARGUMENTS'.split(' ').filter(Boolean);
const depth = args.find(a => a.startsWith('--depth='))?.split('=')[1] || 'medium';
const topic = args.filter(a => !a.startsWith('--')).join(' ');

Input

输入

Arguments:

<topic> [--depth=brief|medium|deep]

topic: Subject to research (required)
--depth: Source gathering depth
- ```
brief
```
  : 10 sources (quick overview)
- ```
medium
```
  : 20 sources (default, balanced)
- ```
deep
```
  : 40 sources (comprehensive)

参数:

<topic> [--depth=brief|medium|deep]

topic: 要调研的主题（必填）
--depth: 资源收集深度
- ```
brief
```
  : 10个来源（快速概览）
- ```
medium
```
  : 20个来源（默认，平衡选项）
- ```
deep
```
  : 40个来源（全面调研）

Research Methodology

调研方法

Based on best practices from:

Anthropic's Context Engineering
DeepLearning.AI Tool Use Patterns
Anara's AI Literature Reviews

基于以下来源的最佳实践：

Anthropic上下文工程
DeepLearning.AI工具使用模式
Anara AI文献综述

1. Progressive Query Architecture

1. 渐进式查询架构

Use funnel approach to avoid noise from long query lists:

Broad Phase (landscape mapping):

"{topic} overview introduction"
"{topic} documentation official"

Focused Phase (core content):

"{topic} best practices"
"{topic} examples tutorial"
"{topic} site:stackoverflow.com"

Deep Phase (advanced, if depth=deep):

"{topic} advanced techniques"
"{topic} pitfalls mistakes avoid"
"{topic} 2025 2026 latest"

使用漏斗式方法避免长查询列表产生的噪声：

广度阶段（全景梳理）:

"{topic} overview introduction"
"{topic} documentation official"

聚焦阶段（核心内容）:

"{topic} best practices"
"{topic} examples tutorial"
"{topic} site:stackoverflow.com"

深度阶段（高阶内容，仅当depth=deep时执行）:

"{topic} advanced techniques"
"{topic} pitfalls mistakes avoid"
"{topic} 2025 2026 latest"

2. Source Quality Scoring

2. 来源质量评分

Multi-dimensional evaluation (max score: 100):

Factor	Weight	Max	Criteria
Authority	3x	30	Official docs (10), recognized expert (8), established site (6), blog (4), random (2)
Recency	2x	20	<6mo (10), <1yr (8), <2yr (6), <3yr (4), older (2)
Depth	2x	20	Comprehensive (10), detailed (8), overview (6), superficial (4), fragment (2)
Examples	2x	20	Multiple code examples (10), one example (6), no examples (2)
Uniqueness	1x	10	Unique perspective (10), some overlap (6), duplicate content (2)

Selection threshold: Top N sources by score (N = depth target)

多维度评估（满分100分）:

评估维度	权重	最高分	评分标准
权威性	3x	30	官方文档 (10), 公认专家产出 (8), 成熟站点内容 (6), 博客 (4), 未知来源内容 (2)
时效性	2x	20	<6个月 (10), <1年 (8), <2年 (6), <3年 (4), 更早 (2)
内容深度	2x	20	内容全面 (10), 内容详细 (8), 概览级 (6), 表面内容 (4), 碎片化内容 (2)
示例丰富度	2x	20	多个代码示例 (10), 1个示例 (6), 无示例 (2)
内容独特性	1x	10	独特视角 (10), 部分重叠 (6), 重复内容 (2)

筛选阈值: 按得分取前N个来源（N为深度对应的目标数量）

3. Just-In-Time Retrieval

3. 即时检索

Don't pre-load all content (causes context rot):

Collect URLs first via WebSearch
Score based on metadata (title, description, URL)
Fetch only selected sources via WebFetch
Extract summaries (not full content)

不要预加载所有内容（会导致上下文失效）:

先通过WebSearch收集URL
基于元数据（标题、描述、URL）评分
仅通过WebFetch拉取选中的来源内容
提取摘要（而非完整内容）

4. Content Extraction Guidelines

4. 内容提取规范

For each source, extract:

json

{
  "url": "https://...",
  "title": "Article Title",
  "qualityScore": 85,
  "scores": {
    "authority": 9,
    "recency": 8,
    "depth": 7,
    "examples": 9,
    "uniqueness": 6
  },
  "keyInsights": [
    "Concise insight 1",
    "Concise insight 2"
  ],
  "codeExamples": [
    {
      "language": "javascript",
      "description": "Basic usage pattern"
    }
  ],
  "extractedAt": "2026-02-05T12:00:00Z"
}

Copyright compliance: Summaries and insights only, never verbatim paragraphs.

对每个来源提取以下信息：

json

{
  "url": "https://...",
  "title": "Article Title",
  "qualityScore": 85,
  "scores": {
    "authority": 9,
    "recency": 8,
    "depth": 7,
    "examples": 9,
    "uniqueness": 6
  },
  "keyInsights": [
    "Concise insight 1",
    "Concise insight 2"
  ],
  "codeExamples": [
    {
      "language": "javascript",
      "description": "Basic usage pattern"
    }
  ],
  "extractedAt": "2026-02-05T12:00:00Z"
}

版权合规要求: 仅提取摘要和洞见，不得直接复制原文段落。

Output Structure

输出结构

Topic Guide Template

主题指南模板

Create

agent-knowledge/{slug}.md

markdown

undefined

创建

agent-knowledge/{slug}.md

markdown

undefined

Learning Guide: {Topic}

学习指南: {Topic}

Generated: {date} Sources: {count} resources analyzed Depth: {brief|medium|deep}

生成时间: {date} 来源: 共分析{count}份资源 调研深度: {brief|medium|deep}

Prerequisites

前置知识

What you should know before diving in:

Prerequisite 1
Prerequisite 2

开始学习前你需要了解：

前置知识1
前置知识2

TL;DR

核心速览

Essential points in 3-5 bullets:

Key point 1
Key point 2
Key point 3

3-5条核心要点：

核心要点1
核心要点2
核心要点3

Core Concepts

核心概念

{Concept 1}

{概念1}

{Synthesized explanation from multiple sources}

Key insight: {Most important takeaway}

{整合多来源内容的解释}

核心洞见: {最重要的结论}

{Concept 2}

{概念2}

{Synthesized explanation}

{整合多来源内容的解释}

Code Examples

代码示例

Basic Example

基础示例

// Description of what this demonstrates
{code}

// 示例功能说明
{code}

Advanced Pattern

高阶用法

{code}

{code}

Common Pitfalls

常见陷阱

Pitfall	Why It Happens	How to Avoid
Issue 1	Root cause	Prevention strategy

陷阱	产生原因	规避方法
问题1	根本原因	预防方案

Best Practices

最佳实践

Synthesized from {n} sources:

Practice 1: Explanation
Practice 2: Explanation

整合自{n}份来源：

实践1: 说明
实践2: 说明

拓展阅读

Resource	Type	Why Recommended
Title	Official Docs	Authoritative reference
Title	Tutorial	Step-by-step guide

Generated by /learn from {count} sources. See
resources/{slug}-sources.json
for full source metadata.

undefined

资源	类型	推荐理由
标题	官方文档	权威参考资料
标题	教程	分步操作指南

由/learn命令基于{count}份来源生成。 完整来源元数据见
resources/{slug}-sources.json
。

undefined

Master Index Template

主索引模板

Create/update

agent-knowledge/CLAUDE.md

markdown

undefined

创建/更新

agent-knowledge/CLAUDE.md

markdown

undefined

Agent Knowledge Base

Agent知识库

Learning guides created by /learn. Reference these when answering questions about listed topics.

由/learn命令创建的学习指南，回答所列主题相关问题时可参考这些内容。

Available Topics

可用主题

Topic	File	Sources	Depth	Created
{Topic 1}	{slug1}.md	{n}	medium	2026-02-05
{Topic 2}	{slug2}.md	{n}	deep	2026-02-04

主题	文件	来源数量	深度	创建时间
{主题1}	{slug1}.md	{n}	medium	2026-02-05
{主题2}	{slug2}.md	{n}	deep	2026-02-04

Trigger Phrases

触发短语

Use this knowledge when user asks about:

"How does {topic1} work?" → {slug1}.md
"Explain {topic1}" → {slug1}.md
"{Topic2} best practices" → {slug2}.md

用户询问以下内容时使用该知识库：

"How does {topic1} work?" → {slug1}.md
"Explain {topic1}" → {slug1}.md
"{Topic2} best practices" → {slug2}.md

Quick Lookup

快速查询

Keyword	Guide
recursion	recursion.md
hooks, react	react-hooks.md

关键词	对应指南
recursion	recursion.md
hooks, react	react-hooks.md

How to Use

使用方法

Check if user question matches a topic
Read the relevant guide file
Answer based on synthesized knowledge
Cite the guide if user asks for sources


Copy to `agent-knowledge/AGENTS.md` for OpenCode/Codex.

检查用户问题是否匹配已有主题
读取对应的指南文件
基于整合后的知识回答
如果用户要求提供来源，引用对应指南


复制到 `agent-knowledge/AGENTS.md` 供OpenCode/Codex使用。

Sources Metadata

来源元数据

Create

agent-knowledge/resources/{slug}-sources.json

json

{
  "topic": "{original topic}",
  "slug": "{slug}",
  "generated": "2026-02-05T12:00:00Z",
  "depth": "medium",
  "totalSources": 20,
  "sources": [
    {
      "url": "https://...",
      "title": "...",
      "qualityScore": 85,
      "scores": {
        "authority": 9,
        "recency": 8,
        "depth": 7,
        "examples": 9,
        "uniqueness": 6
      },
      "keyInsights": ["..."]
    }
  ]
}

创建

agent-knowledge/resources/{slug}-sources.json

json

{
  "topic": "{original topic}",
  "slug": "{slug}",
  "generated": "2026-02-05T12:00:00Z",
  "depth": "medium",
  "totalSources": 20,
  "sources": [
    {
      "url": "https://...",
      "title": "...",
      "qualityScore": 85,
      "scores": {
        "authority": 9,
        "recency": 8,
        "depth": 7,
        "examples": 9,
        "uniqueness": 6
      },
      "keyInsights": ["..."]
    }
  ]
}

Self-Evaluation Checklist

自检清单

Before finalizing, rate output (1-10):

Metric	Question	Target
Coverage	Does guide cover main aspects?	≥7
Diversity	Are sources from diverse types?	≥6
Examples	Are code examples practical?	≥7
Accuracy	Confidence in content accuracy?	≥8

Flag gaps: Note any important subtopics not covered.

最终输出前按1-10分评分：

指标	评分问题	目标分
覆盖度	指南是否覆盖了主题的主要方面？	≥7
来源多样性	来源是否来自多种不同类型？	≥6
示例实用性	代码示例是否实用？	≥7
内容准确率	对内容准确率的信心？	≥8

标记缺口: 记录所有未覆盖的重要子主题。

Enhancement Integration

增强功能集成

If enhance=true, invoke after guide creation:

javascript

// Enhance the topic guide for RAG
Skill({ name: 'enhance-docs', args: `agent-knowledge/${slug}.md --ai` });

// Enhance the master index
Skill({ name: 'enhance-prompts', args: 'agent-knowledge/CLAUDE.md' });

如果enhance=true，在指南创建完成后调用：

javascript

// 为RAG优化主题指南
Skill({ name: 'enhance-docs', args: `agent-knowledge/${slug}.md --ai` });

// 优化主索引
Skill({ name: 'enhance-prompts', args: 'agent-knowledge/CLAUDE.md' });

Output Format

输出格式

Return structured JSON between markers:

=== LEARN_RESULT ===
{
  "topic": "recursion",
  "slug": "recursion",
  "depth": "medium",
  "guideFile": "agent-knowledge/recursion.md",
  "sourcesFile": "agent-knowledge/resources/recursion-sources.json",
  "sourceCount": 20,
  "sourceBreakdown": {
    "officialDocs": 4,
    "tutorials": 5,
    "stackOverflow": 3,
    "blogPosts": 5,
    "github": 3
  },
  "selfEvaluation": {
    "coverage": 8,
    "diversity": 7,
    "examples": 9,
    "accuracy": 8,
    "gaps": ["tail recursion optimization not covered"]
  },
  "enhanced": true,
  "indexUpdated": true
}
=== END_RESULT ===

在标记之间返回结构化JSON：

=== LEARN_RESULT ===
{
  "topic": "recursion",
  "slug": "recursion",
  "depth": "medium",
  "guideFile": "agent-knowledge/recursion.md",
  "sourcesFile": "agent-knowledge/resources/recursion-sources.json",
  "sourceCount": 20,
  "sourceBreakdown": {
    "officialDocs": 4,
    "tutorials": 5,
    "stackOverflow": 3,
    "blogPosts": 5,
    "github": 3
  },
  "selfEvaluation": {
    "coverage": 8,
    "diversity": 7,
    "examples": 9,
    "accuracy": 8,
    "gaps": ["tail recursion optimization not covered"]
  },
  "enhanced": true,
  "indexUpdated": true
}
=== END_RESULT ===

Error Handling

错误处理

Error	Action
WebSearch fails	Retry with simpler query
WebFetch timeout	Skip source, note in metadata
<minSources found	Warn user, proceed with available
Enhancement fails	Skip, note in output
Index doesn't exist	Create new index

错误	处理动作
WebSearch失败	使用更简单的查询重试
WebFetch超时	跳过该来源，在元数据中记录
找到的来源少于最低要求	警告用户，使用现有资源继续
增强功能调用失败	跳过，在输出中记录
索引不存在	创建新索引

Token Budget

Token预算

Estimated token usage by phase:

Phase	Tokens	Notes
WebSearch queries	~2,000	5-8 queries
Source scoring	~1,000	Metadata only
WebFetch extraction	~40,000	20 sources × 2,000 avg
Synthesis	~10,000	Guide generation
Enhancement	~5,000	Two skill calls
Total	~60,000	Within opus budget

各阶段预估token使用量：

阶段	Token数量	说明
WebSearch查询	~2,000	5-8次查询
来源评分	~1,000	仅处理元数据
WebFetch内容提取	~40,000	20个来源 × 平均每个2000token
内容整合	~10,000	指南生成
增强处理	~5,000	两次Skill调用
总计	~60,000	在opus模型预算范围内

Integration

集成说明

This skill is invoked by:

```
learn-agent
```
for
```
/learn
```
command
Potentially other research-oriented agents

该Skill可由以下模块调用：

```
/learn
```
命令对应的
```
learn-agent
```
其他面向调研的Agent也可调用

learn

Original

Translation

learn

learn

Parse Arguments

参数解析

Input

输入

Research Methodology

调研方法

1. Progressive Query Architecture

1. 渐进式查询架构

2. Source Quality Scoring

2. 来源质量评分

3. Just-In-Time Retrieval

3. 即时检索

4. Content Extraction Guidelines

4. 内容提取规范

Output Structure

输出结构

Topic Guide Template

主题指南模板

Learning Guide: {Topic}

学习指南: {Topic}

Prerequisites

前置知识

TL;DR

核心速览

Core Concepts

核心概念

{Concept 1}

{概念1}

{Concept 2}

{概念2}

Code Examples

代码示例

Basic Example

基础示例

Advanced Pattern

高阶用法

Common Pitfalls

常见陷阱

Best Practices

最佳实践

Further Reading

拓展阅读

Master Index Template

主索引模板

Agent Knowledge Base

Agent知识库

Available Topics

可用主题

Trigger Phrases

触发短语

Quick Lookup

快速查询

How to Use

使用方法

Sources Metadata

来源元数据

Self-Evaluation Checklist

自检清单

Enhancement Integration

增强功能集成

Output Format

输出格式

Error Handling

错误处理

Token Budget

Token预算

Integration

集成说明