Best Practice Scout

技术决策时找行业最佳实践 + 跨学科深刻洞见。用于选型、架构决策、未知领域问题;触发词「最佳实践」「业界怎么做」「有类似的吗」「best practice」「cross-domain」「跨学科启发」。

107 stars 21 forksOther

Install

skills CLI (Claude Code, Cursor, Codex and more)
npx skills add Dimon94/skills --skill best-practice-scout
Manual install
git clone --depth 1 https://github.com/Dimon94/skills.git /tmp/skills
mkdir -p ~/.claude/skills
cp -r /tmp/skills/skills/best-practice-scout ~/.claude/skills/best-practice-scout

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# Best Practice Scout

技术决策时找到行业最佳实践和跨学科的**垂直深度洞见**。

## 核心原则

跨学科映射不是为了类比而类比,必须带来**本行业没有的深刻见解**:
1. 不是平行类比("并发 ≈ 交通"),而是**垂直深挖**(针对具体盲点的精细化机制)
2. 必须落到**可执行的细节**(算法/模型/决策树,不是原则)
3. 必须指出**本行业现有方案的盲点**,跨学科方案填补了什么空白

## Step 1. 本地语境

用 `investigate` 理解问题在代码库中的位置和约束。

**完成标准**:能用一句话说清"解决什么问题、在什么约束下"。

## Step 2. 行业实践 + 盲点识别

启动一个或多个 `research` 子 agent(一个技术栈一个 agent):

### 任务
1. 搜索本技术栈同类问题的主流解法
2. 优先 official docs、production cases、ADRs
3. **识别共同局限**:所有方案都假设什么?都没覆盖什么场景?

### 输出
- 3-5 个主流方案 + 各自 tradeoffs
- **至少 1 个具体的、可验证的盲点**

**完成标准**:每个方案带来源 + 至少 1 条 tradeoff;盲点必须是**可验证的具体局限**(不是"可能不够灵活"这种模糊表述)。

## Step 3. 跨学科深挖(针对盲点)

**等 Step 2 完成后**,针对每个识别出的盲点,启动一个或多个 `research` 子 agent(一个学科/领域一个 agent)。

### 任务
1. **输入**:一个具体盲点
2. **搜索**:哪个学科有**专门处理这个局限**的成熟框架/模型/算法
3. **提取**:可迁移的**具体机制**(不是"参考思路")

### 输出
每个跨学科方案必须包含:
1. **盲点定位**:本行业方案的共同局限是什么
2. **深度机制**:该学科中针对这个局限的**精细化框架/模型/算法**(带术语、流程、公式)
3. **可迁移机制**:这个机制可以**直接用在当前问题的哪个环节**
4. **差异化价值**:为什么本行业没有(本行业的假设/约束导致这个盲点)

**完成标准**:
- 不允许平行类比("X 和 Y 都有资源调度")
- 必须给出**可执行机制**(算法步骤/决策树/公式)
- 必须说明**本行业为什么没有**
- **新颖性核实**:用该机制的本行业说法回搜本行业(文档、论文、主流实现)。搜到已有实现就降级为「本行业已有」,附来源;没搜到就写明搜索词与范围,结论标「推断」
- 跨学科术语与机制都要有来源引用,不凭记忆举例

### 示例对比

❌ **空泛映射**:
"并发控制和交通管理都涉及资源调度,可以参考交通管理的思路。"

❌ **未核实的「深度」映射**(术语真实,新颖性不成立):
"交通管理的**动态车道分配**(FHWA:按路况动态开关单条车道并提前预警)和可逆车道切换的**清空间隔**(MUTCD:黄 X → 红 X,等车辆驶离再放对向进入)可迁移为:高竞争时动态调整锁粒度。现有并发方案没有这个维度。"

新颖性核实否决了最后一句:数据库早已有锁升级,竞争高时也会降级到更细的锁(SQL Server 锁升级文档、Silberschatz《Database System Concepts》)。正确产出是「本行业已有」,不是新洞见。
Read the full skill on GitHub

Details

Skill name
best-practice-scout
Repository
Dimon94/skills
Path
skills/best-practice-scout/SKILL.md
Repo updated
18 November 2026

FAQ

What does the Best Practice Scout skill do?

技术决策时找行业最佳实践 + 跨学科深刻洞见。用于选型、架构决策、未知领域问题;触发词「最佳实践」「业界怎么做」「有类似的吗」「best practice」「cross-domain」「跨学科启发」。

How do I install the Best Practice Scout skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add Dimon94/skills --skill best-practice-scout. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Best Practice Scout popular?

It lives in Dimon94/skills, which has 107 GitHub stars and 21 forks.

Related skills

Maintain Verification Skill

107

Dimon94/skills

Periodic pass that keeps a project's verification skill and feature map honest: parallel source readers per feature, one live session driving every feature, at most one PR of proven corrections. Use for /maintain-verification-skill or "audit the verify skill".

Git & Code ReviewIn 2 lists21

Create Verification Skill

107

Dimon94/skills

Generate a project-local verification skill that drives your app the way a user does — any language, framework, or platform. Use for /create-verification-skill, "make a control skill for this repo", or when a project has no scripted way to prove UI/CLI/service behavior.

Software DevelopmentIn 2 lists21

How

107

Dimon94/skills

Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which package owns this", "is this the right layer"). Explains subsystem architecture, runtime flow, onboarding mental models. Use why for motivation.

Software DevelopmentIn 2 lists21

Better Colors

107

Dimon94/skills

Helps you build a color system and answer anything about color in your project. You can generate palettes, use semantic tokens, convert between formats, check contrast and more.

Other21

Benchmark Checklist

107

Dimon94/skills

Vet a perf measurement (limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened) before you report or act on it. Use when you run a benchmark or report a speedup or regression you measured.

Other21

Better Layout

107

Dimon94/skills

Helps with grouping, alignment, reading order, progressive disclosure and other details that make a good layout.

Other21

Browse the most popular Claude skills, all other skills, or write your own with the skills generator.