Autopolio

Turn your GitHub into a professional portfolio — powered by AI

🇰🇷 한국어 README

Autopolio is an open-source, AI-powered portfolio and resume automation platform. It analyzes your GitHub repositories, builds a structured career knowledge base, and generates tailored resumes for major job platforms — all without manually writing every line.

---

Screenshots

Dashboard & Career Timeline	Project Management
Job Platform Templates	AI Provider Settings

Additional screenshots: Document Generation · Template Management · GitHub Setup

---

Why Autopolio

Most developers have rich GitHub histories but blank resume pages. Writing a tailored resume for each job platform (Saramin, Remember, Jumpit) takes hours of repetitive effort. Autopolio closes that gap: connect your GitHub, define your career history once, and let the AI pipeline do the rest.

---

Key Features

• GitHub Repository Analysis — Parses commit history and auto-detects 200+ technologies across JavaScript, Python, Java, Kotlin, Dart, PHP, and more — no LLM required for detection
• Per-Contributor Breakdown — Isolates your commits from team repos; Conventional Commit parsing, work area detection, and code quality metrics
• Career Knowledge Base — Structured management of companies, projects, achievements, certifications, education, and awards
• Platform Resume Templates — Pre-built HTML templates for Saramin, Remember, and Jumpit; Mustache syntax for fully custom templates
• Multi-LLM AI Summarization — OpenAI GPT-4, Anthropic Claude, and Google Gemini all supported; switchable per request
• Multiple Export Formats — DOCX, PDF, and Markdown output from a single source
• Electron Desktop App — Cross-platform installer for Windows (exe), macOS (dmg), and Linux (AppImage) with local CLI tool support
• Internationalization — Full UI in Korean and English via react-i18next
• CI/CD Pipeline — GitHub Actions with pytest, Playwright, ruff, tsc, Bandit security scanning, and Gemini Code Assist review

---

How It Works

Step 1 — GitHub Analysis       Fetch commits and extract statistics (parallel)
Step 2 — Code Extraction       Detect code patterns and project architecture
Step 3 — Tech Detection        Auto-detect tech stack from dependency files (fast, no LLM)
Step 4 — Achievement Detection Extract quantitative achievements from commit messages
Step 5 — LLM Summarization     Generate AI-powered project summaries (parallel)
Step 6 — Template Mapping      Map structured data to platform-specific template fields
Step 7 — Document Generation   Produce DOCX / PDF / Markdown output

---

Quick Start

Option A: Docker (recommended)

git clone https://github.com/sehoon787/autopolio.git
cd Autopolio

cp .env.example .env
# Edit .env — set ENCRYPTION_KEY, GITHUB_CLIENT_ID/SECRET, and one LLM API key

docker-compose up -d

Open http://localhost:3035 in your browser.

Option B: Dev script

git clone https://github.com/sehoon787/autopolio.git
cd Autopolio
cp .env.example .env

# Windows
start-dev.bat

# Linux / macOS
./start-dev.sh

Option C: Manual

# Backend
uv sync
uv run uvicorn api.main:app --reload --port 8085

# Frontend (separate terminal)
cd frontend
npm install
npm run dev

Service	URL
Frontend	http://localhost:3035
API Docs (Swagger)	http://localhost:8085/docs
API Docs (ReDoc)	http://localhost:8085/redoc

Ports are configured in config/runtime.yaml.

---

Electron Desktop App

Run Autopolio as a standalone desktop application with built-in backend:

Development

# 1. Start the backend
uv sync
uv run uvicorn api.main:app --reload --port 8085

# 2. Start Electron (separate terminal)
cd frontend
npm install
npm run electron:dev

Build Installers

cd frontend

npm run electron:build:win      # Windows exe (NSIS)
npm run electron:build:mac      # macOS dmg
npm run electron:build:linux    # Linux AppImage

The desktop app auto-detects CLI tools (Claude Code, Gemini CLI) and manages API keys locally.

---

Testing

# Full suite: Docker → pytest → Playwright
tests/scripts/run-all.sh          # Linux / macOS
tests/scripts/run-all.bat         # Windows

# API tests only
tests/scripts/run-api-tests.sh

# E2E tests only
tests/scripts/run-e2e-tests.sh

Seed Data

Populate a local environment with realistic sample data:

python tests/seed_sample_data.py            # Insert sample data
python tests/seed_sample_data.py --clean    # Wipe first, then insert
python tests/seed_sample_data.py --create-user  # Create user + insert

Inserts 3 companies, 6 projects, 2 education records, 4 training records, 3 certifications, 2 awards, 3 publications, 2 patents, and 2 activities.

---

Tech Stack

Layer	Technologies
Backend	FastAPI, SQLAlchemy, SQLite
Frontend	React 19, TypeScript, Vite, Tailwind CSS, Shadcn/ui
Desktop	Electron, electron-builder, electron-serve
State Management	Zustand, TanStack Query
Internationalization	react-i18next, i18next
LLM	OpenAI GPT-4 / Anthropic Claude / Google Gemini
Document Generation	python-docx, reportlab, chevron (Mustache)
Package Managers	uv (Python), npm (Frontend)
CI/CD	GitHub Actions + Gemini Code Assist
Testing	pytest, Playwright

---

Project Structure

Autopolio/
├── api/                     # FastAPI backend
│   ├── constants/           # Centralized enums and config constants
│   ├── models/              # SQLAlchemy ORM models
│   ├── schemas/             # Pydantic request/response schemas
│   ├── routers/             # API route handlers (modularized)
│   └── services/            # Business logic (modularized)
├── frontend/                # React frontend + Electron
│   ├── electron/            # Electron main process and services
│   └── src/
│       ├── api/             # API client functions
│       ├── components/      # Reusable UI components
│       ├── locales/         # i18n translation files (ko, en)
│       ├── pages/           # Page components
│       └── stores/          # Zustand state stores
├── .github/workflows/       # CI/CD pipeline definitions
├── tests/                   # Test scripts and E2E tests
├── config/                  # YAML configuration files
├── data/                    # SQLite database and platform HTML templates
├── docs/                    # Project documentation and screenshots
├── result/                  # Generated output documents
├── pyproject.toml           # Python dependencies (uv)
└── docker-compose.yml

---

Key API Endpoints

Method	Path	Description
POST	/api/users	Create a user
GET	/api/github/repos	List connected repositories
POST	/api/github/analyze	Start repository analysis
GET	/api/github/contributor-analysis/{id}	Per-user detailed analysis
GET	/api/knowledge/projects	List projects
GET	/api/platforms	List platform resume templates
POST	/api/platforms/{id}/render	Render a template with user data
POST	/api/pipeline/run	Run the full generation pipeline
GET	/api/documents	List generated documents

Full interactive documentation is at http://localhost:8085/docs when the server is running.

---

Environment Variables

Variable	Description	Required
ENCRYPTION_KEY	Fernet encryption key for stored tokens	Yes
LLM_PROVIDER	Default LLM provider (openai / anthropic / gemini)	Yes
OPENAI_API_KEY	OpenAI API key	One of the three
ANTHROPIC_API_KEY	Anthropic API key	One of the three
GEMINI_API_KEY	Google Gemini API key	One of the three
GITHUB_CLIENT_ID	GitHub OAuth App client ID	Yes
GITHUB_CLIENT_SECRET	GitHub OAuth App client secret	Yes
DATABASE_URL	SQLite database path (default: data/autopolio.db)	No

---

Contributing

Contributions are welcome. See CONTRIBUTING.md for guidelines on reporting bugs, proposing features, and submitting pull requests.

This project follows the Contributor Covenant Code of Conduct.

License

Apache License 2.0 — see the LICENSE file for details.