PostGen

I have reacehd a point in my life where I want to vibe code an application for every task that bothers me. So this again is born from such urge. I have a lot of ideas that I want to post on linkedIn but I am very lazy to write those posts, generate image from Gemini and then upload them. So, I have made this app, so that I can just put my idea in messy words, it will craft a linkedIn post for you.

This app should cost you, but I won't. If you have your own APIs, then go ahead and put your API keys and generate your images. There is no SaaS stuff going on here.

Link: postgen.iashyam.in

One question I am wondering: should I make this open-source, or put a pricing on it? If I make it open-source, someone else will pick my ready-made product and make money from it. And if I put the link [demo] on it, other people will just use that demo instead of cloning from it and I will have to pay for tokens. If I put money-based system, well that's not something I want to do..

Tech Stack

Frontend: Next.js, TypeScript, Tailwind CSS v4, Zustand, TipTap editor

Backend: FastAPI, LangGraph, LangChain + Google Gemini, Motor (async MongoDB), Pillow

Infrastructure: MongoDB, AWS S3 (image storage)

Features

• AI post generation powered by Google Gemini via LangGraph
• Rich text editing with TipTap
• Draft management (save, edit, delete)
• Post history tracking
• LinkedIn OAuth integration for direct publishing
• Image generation and S3 storage
• Per-user settings and preferences

Prerequisites

• Python 3.13+ and uv
• Node.js 18+ and Yarn
• Google API key (Gemini)
• AWS credentials (S3, for image storage)
• LinkedIn OAuth app (for publishing)

Quick Start

# 1. Clone and setup
git clone https://github.com/iashyam/PostGen.git && cd PostGen
make setup

# 2. Configure environment
#    Edit backend/.env with your API keys

# 3. Start database + dev servers
make dev

Backend runs at http://localhost:8000, frontend at http://localhost:3000.

Environment Variables

Copy backend/.env.example to backend/.env and fill in:

Variable	Description
GOOGLE_API_KEY	Google AI / Gemini API key
GEMINI_MODEL	Model name (default: gemini-2.0-flash)
MONGODB_URI	MongoDB connection string
AWS_ACCESS_KEY_ID	AWS access key for S3
AWS_SECRET_ACCESS_KEY	AWS secret key
S3_BUCKET	S3 bucket name for images
LINKEDIN_CLIENT_ID	LinkedIn OAuth client ID
LINKEDIN_CLIENT_SECRET	LinkedIn OAuth client secret
SECRET_KEY	JWT signing secret
FRONTEND_URL	Frontend URL for CORS (default: http://localhost:3000)
APP_URL	Public URL (e.g. Cloudflare tunnel domain) for CORS

Make Commands

Command	Description
make setup	Install deps + create .env from example
make dev	Start backend and frontend
make dev-backend	Start backend only
make dev-frontend	Start frontend only
make build	Build frontend for production
make lint	Run TypeScript type check
make clean	Remove build artifacts and caches

Project Structure

PostGen/
├── backend/
│   └── app/
│       ├── main.py              # FastAPI app entrypoint
│       ├── config.py            # Pydantic settings
│       ├── database.py          # MongoDB connection
│       ├── models/              # Pydantic models (user, post, draft, settings)
│       ├── routers/             # API routes (auth, generate, drafts, history, linkedin, settings)
│       ├── services/            # Business logic (langgraph_chain, linkedin, s3, image)
│       └── utils/
├── frontend/
│   └── src/
│       ├── pages/               # CreatePost, Drafts, History, Settings
│       ├── components/          # UI components by feature
│       ├── store/               # Zustand state management
│       ├── api/                 # API client
│       ├── hooks/
│       └── types/
├── docker-compose.yml
└── Makefile

API

Health check: GET /api/health

Core endpoints under /api:

• POST /generate — AI post generation
• GET/POST/PUT/DELETE /drafts — Draft CRUD
• GET /history — Post history
• POST /linkedin/publish — Publish to LinkedIn
• GET/PUT /settings — User settings

Auth endpoints under /api/auth:

• LinkedIn OAuth flow

License

MIT