Language: This skill is written in English for universal compatibility. Always respond in the user's language unless explicitly requested otherwise. If uncertain about the user's language, ask for clarification.

Overview Workflow

Generate a concise 60-80 line project overview from existing documentation.

⚠️ Critical Constraint

OUTPUT MUST BE 80-120 LINES MAXIMUM. (excluding mermaid code blocks)

This is a "5-minute read" document, NOT a detailed spec.

•Do NOT copy sections from arch.md
•Do NOT list all API endpoints
•Do NOT list all data/entities
•Just summarize and link to detailed docs
•DO include visual diagrams (Architecture, ERD, Key Flow)

💡 Recommended Model

Sonnet recommended (summarization task)

📁 Document Structure

Input:

code

projectRoot/
  └── docs/
        └── {serviceName}/
              ├── spec.md  # Source
              └── arch.md     # Source

Output:

code

projectRoot/
  └── docs/
        └── {serviceName}/
              └── overview.md      # Generated

Phase 0: Skill Entry

0-1. Collect Service Name

Use AskQuestion:

json

{
  "title": "Project Overview Generator",
  "questions": [
    {
      "id": "service_name",
      "prompt": "Which service's overview do you want to generate?",
      "options": [
        {"id": "input", "label": "I will type the service name"}
      ]
    }
  ]
}

0-2. Load Source Documents

Read:

•docs/{serviceName}/spec.md
•docs/{serviceName}/arch.md

If not found → Error: "Source documents not found. Run /spec and /arch first."

Phase 1: Auto Extract

Extract from existing documents:

From spec.md

Field	Location
Project Name	Header / Service Name
One-liner Description	Service Description section
Key Features	Functional Requirements
Target Users	User definition section

From arch.md

Field	Location
Tech Stack	Tech Stack section
Architecture Diagram	System Architecture section
Database Overview	DB Schema (table count, main entities)
ERD (main entities)	DB Schema → Extract 4-6 core entities and relationships
Key Sequence Flow	Sequence Diagram → Pick most important user flow
API Overview	API Spec (endpoint count, categories)
External Integrations	External API section

Phase 2: Additional Info (AskQuestion)

Ask user for info not in design docs:

json

{
  "title": "Additional Information",
  "questions": [
    {
      "id": "repo_url",
      "prompt": "Repository URL? (leave empty if N/A)",
      "options": [
        {"id": "github", "label": "GitHub"},
        {"id": "gitlab", "label": "GitLab"},
        {"id": "other", "label": "Other (I will paste)"},
        {"id": "skip", "label": "Skip"}
      ]
    },
    {
      "id": "demo_url",
      "prompt": "Demo/Staging URL? (leave empty if N/A)",
      "options": [
        {"id": "has_url", "label": "Yes, I will provide"},
        {"id": "skip", "label": "Skip (not deployed yet)"}
      ]
    },
    {
      "id": "local_setup",
      "prompt": "How to run locally?",
      "options": [
        {"id": "docker", "label": "Docker Compose"},
        {"id": "manual", "label": "Manual setup (I will describe)"},
        {"id": "skip", "label": "Skip for now"}
      ]
    },
    {
      "id": "contact",
      "prompt": "Team contact or maintainer?",
      "options": [
        {"id": "provide", "label": "I will provide"},
        {"id": "skip", "label": "Skip"}
      ]
    }
  ]
}

Phase 3: Generate overview.md

Template Structure (80-120 lines MAX, excluding mermaid blocks)

markdown

# {Project Name}

> {One-liner description - 1 sentence}

## At a Glance

| Item | Value |
|------|-------|
| Status | 🟢 Active / 🟡 Development / 🔴 Deprecated |
| Tech Stack | {main 3-4 technologies only} |
| Repository | [Link]({repo_url}) |

## What is this?

{2-3 sentences ONLY. What + Why + For whom}

## Key Features

- ✅ {Feature 1}
- ✅ {Feature 2}
- ✅ {Feature 3}
- ✅ {Feature 4}
- ✅ {Feature 5}

(MAX 5-6 features. No sub-lists.)

## Architecture

```mermaid
graph LR
    {SIMPLIFIED diagram - max 6-8 nodes}

Layer	Technology
Backend	{1 line}
Database	{1 line}
Infra	{1 line}

(MAX 4-5 rows. No detailed breakdown.)

Data Model

mermaid

erDiagram
    {Main entities only - max 4-6 entities}
    {Show key relationships}

(Only core entities. Link to arch.md for full schema.)

Key Flow

mermaid

sequenceDiagram
    {Most important user flow - max 6-8 steps}

(Pick ONE key flow. Link to arch.md for all flows.)

Quick Start

bash

git clone {repo}
cd {project}
docker-compose up -d  # or equivalent 2-3 commands

Learn More

Doc	What's inside
Requirements	기능 요구사항
Architecture	상세 설계, API 명세, DB 스키마

Generated by archflow /overview

code


### What NOT to include

❌ API endpoint lists → Link to arch.md
❌ DB schema details → Link to arch.md
❌ Data coverage lists → Link to arch.md
❌ Detailed prerequisites → Link to README or arch.md
❌ Multiple setup methods → Pick one, simplest

---

## Phase 4: Write & Confirm

1. Write `docs/{serviceName}/overview.md`
2. Show preview to user
3. Ask for any adjustments

```json
{
  "title": "Overview Generated",
  "questions": [
    {
      "id": "satisfaction",
      "prompt": "Overview generated. Any changes needed?",
      "options": [
        {"id": "done", "label": "Looks good, done!"},
        {"id": "edit", "label": "Need some edits (I will describe)"},
        {"id": "regenerate", "label": "Regenerate with different focus"}
      ]
    }
  ]
}

Completion Message

✅ Overview Generated

Output: docs/{serviceName}/overview.md

This document provides:

•Project introduction for new team members

•Quick reference for stakeholders

•Getting started guide for developers

Tip: Keep this updated when major changes happen.

Output Example (Target: ~100 lines, excluding mermaid blocks)

markdown

# 투자일기 (Invest Diary)

> AI가 내 투자 판단을 분석해주는 개인 투자일기 시스템

## At a Glance

| Item | Value |
|------|-------|
| Status | 🟡 Development (MVP) |
| Tech Stack | FastAPI, PostgreSQL, Redis, GPT-4 |
| Repository | github.com/user/invest-diary |

## What is this?

투자일기를 작성하면 AI가 과거 투자 판단의 정확도를 분석해줍니다.
예측 정확도, 논리 일관성, 리스크 인식을 평가하여 투자 실력 향상을 돕습니다.

## Key Features

- ✅ Google OAuth 소셜 로그인
- ✅ 포트폴리오 관리 (한국/미국 주식)
- ✅ 투자일기 작성 + 매매기록
- ✅ AI 기반 투자 판단 분석 (GPT-4)
- ✅ 실시간 분석 결과 (SSE)

## Architecture

```mermaid
graph LR
    Client --> API[FastAPI]
    API --> DB[(PostgreSQL)]
    API --> Cache[(Redis)]
    API --> Worker[Celery]
    Worker --> LLM[GPT-4]

Layer	Technology
Backend	FastAPI + SQLAlchemy
Database	PostgreSQL, Redis
AI	OpenAI GPT-4

Data Model

mermaid

erDiagram
    User ||--o{ Portfolio : owns
    User ||--o{ Diary : writes
    Portfolio ||--o{ Stock : contains
    Diary ||--o{ Trade : records
    Diary ||--|| Analysis : has

Key Flow

mermaid

sequenceDiagram
    participant U as User
    participant A as API
    participant W as Worker
    participant L as GPT-4

    U->>A: POST /diary (작성)
    A->>A: Save to DB
    A->>W: Queue analysis
    A-->>U: 202 Accepted
    W->>L: Analyze diary
    L-->>W: Analysis result
    W->>A: SSE push
    A-->>U: Real-time update

Quick Start

bash

git clone <repo> && cd invest-diary
cp .env.example .env  # Edit API keys
docker-compose up -d

Learn More

Doc	What's inside
Requirements	기능 요구사항, 분석 프레임워크
Architecture	API 명세, DB 스키마, 시퀀스 다이어그램

Generated by archflow /overview

code


**Line count: ~85 lines (excluding mermaid code blocks)** ✅