understand-domain

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

/understand-domain

Extracts business domain knowledge — domains, business flows, and process steps — from a codebase and produces an interactive horizontal flow graph in the dashboard.

从代码库中提取业务领域知识——包括领域、业务流程和处理步骤——并在仪表板中生成交互式水平流程图。

How It Works

工作原理

If a knowledge graph already exists (
```
.understand-anything/knowledge-graph.json
```
), derives domain knowledge from it (cheap, no file scanning)
If no knowledge graph exists, performs a lightweight scan: file tree + entry point detection + sampled files
Use
```
--full
```
flag to force a fresh scan even if a knowledge graph exists

如果已存在知识图谱（
```
.understand-anything/knowledge-graph.json
```
），则基于该图谱推导领域知识（成本低，无需扫描文件）
如果不存在知识图谱，则执行轻量扫描：文件树+入口点检测+抽样文件
使用
```
--full
```
flag可强制执行全新扫描，即便已有知识图谱也不例外

Instructions

使用说明

Phase 1: Detect Existing Graph

阶段1：检测已有图谱

Check if

.understand-anything/knowledge-graph.json

exists in the current project

If it exists AND
```
--full
```
was NOT passed → proceed to Phase 3 (derive from graph)
Otherwise → proceed to Phase 2 (lightweight scan)

检查当前项目中是否存在

.understand-anything/knowledge-graph.json

如果文件存在且未传入
```
--full
```
参数 → 进入阶段3（基于图谱推导）
否则 → 进入阶段2（轻量扫描）

Phase 2: Lightweight Scan (Path 1)

阶段2：轻量扫描（路径1）

The preprocessing script does NOT produce a domain graph — it produces raw material (file tree, entry points, exports/imports) so the domain-analyzer agent can focus on the actual domain analysis instead of spending dozens of tool calls exploring the codebase. Think of it as a cheat sheet: cheap Python preprocessing → expensive LLM gets a clean, small input → better results for less cost.

Run the preprocessing script bundled with this skill:
```
python ./extract-domain-context.py <project-root>
```
This outputs
```
<project-root>/.understand-anything/intermediate/domain-context.json
```
containing:
- File tree (respecting
```
.gitignore
```
  )
- Detected entry points (HTTP routes, CLI commands, event handlers, cron jobs, exported handlers)
- File signatures (exports, imports per file)
- Code snippets for each entry point (signature + first few lines)
- Project metadata (package.json, README, etc.)
Read the generated
```
domain-context.json
```
as context for Phase 4
Proceed to Phase 4

预处理脚本不会生成领域图谱，它会生成原始素材（文件树、入口点、导出/导入项），这样domain-analyzer agent就能聚焦于实际的领域分析，而不需要花费几十次工具调用探索代码库。你可以把它看作一份备忘录：低成本的Python预处理 → 昂贵的LLM获得干净、体量小的输入 → 成本更低，效果更好。

运行该skill内置的预处理脚本：
```
python ./extract-domain-context.py <project-root>
```
命令会输出
```
<project-root>/.understand-anything/intermediate/domain-context.json
```
，包含以下内容：
- 文件树（遵循
```
.gitignore
```
  规则）
- 检测到的入口点（HTTP路由、CLI命令、事件处理器、定时任务、导出的处理器）
- 文件特征（每个文件的导出、导入项）
- 每个入口点的代码片段（特征定义+前几行代码）
- 项目元数据（package.json、README等）
读取生成的
```
domain-context.json
```
作为阶段4的上下文
进入阶段4

Phase 3: Derive from Existing Graph (Path 2)

阶段3：基于已有图谱推导（路径2）

Read

.understand-anything/knowledge-graph.json

Format the graph data as structured context:
- All nodes with their types, names, summaries, and tags
- All edges with their types (especially
```
calls
```
  ,
```
imports
```
  ,
```
contains
```
  )
- All layers with their descriptions
- Tour steps if available
This is the context for the domain analyzer — no file reading needed
Proceed to Phase 4

读取

.understand-anything/knowledge-graph.json

将图谱数据格式化为结构化上下文：
- 所有节点及其类型、名称、摘要和标签
- 所有边及其类型（尤其是
```
calls
```
  、
```
imports
```
  、
```
contains
```
  ）
- 所有分层及其描述
- 导览步骤（如果有）
以上内容就是领域分析器的上下文，无需读取任何文件
进入阶段4

Phase 4: Domain Analysis

阶段4：领域分析

Read the domain-analyzer agent prompt from
```
agents/domain-analyzer.md
```
Dispatch a subagent with the domain-analyzer prompt + the context from Phase 2 or 3

The agent writes its output to

.understand-anything/intermediate/domain-analysis.json

从
```
agents/domain-analyzer.md
```
读取domain-analyzer agent的提示词
分配子agent，传入domain-analyzer提示词+阶段2或阶段3生成的上下文

Agent会将输出写入

.understand-anything/intermediate/domain-analysis.json

Phase 5: Validate and Save

阶段5：验证与保存

Read the domain analysis output
Validate using the standard graph validation pipeline (the schema now supports domain/flow/step types)
If validation fails, log warnings but save what's valid (error tolerance)
Save to
```
.understand-anything/domain-graph.json
```

Clean up

.understand-anything/intermediate/domain-analysis.json

and

.understand-anything/intermediate/domain-context.json

读取领域分析输出
使用标准图谱验证管道进行验证（当前schema已支持domain/flow/step类型）
如果验证失败，记录警告但保存所有有效内容（容错机制）
保存到
```
.understand-anything/domain-graph.json
```

清理

.understand-anything/intermediate/domain-analysis.json

和

.understand-anything/intermediate/domain-context.json

Phase 6: Launch Dashboard

阶段6：启动仪表板

Auto-trigger
```
/understand-dashboard
```
to visualize the domain graph
The dashboard will detect
```
domain-graph.json
```
and show the domain view by default

自动触发
```
/understand-dashboard
```
可视化领域图谱
仪表板会检测
```
domain-graph.json
```
，默认展示领域视图