XAI Social OS

AI-operated X/Twitter content engine for SaaS founders and technical creators.

The project is scaffolded as a production-shaped monorepo:

• apps/web - Next.js App Router dashboard with real routes, Tailwind, and shadcn-style components.
• apps/api - FastAPI backend with AI generation, scheduling, worker jobs, and X publishing services.
• docs - architecture and setup notes.

Quick Start

cp .env.example apps/api/.env
cp apps/web/.env.local.example apps/web/.env.local
npm install
npm run dev:web

In another terminal:

cd apps/api
python3 -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
cd ../..
npm run dev:api

The app works in mock-safe mode without API keys. Add OPENROUTER_API_KEY, DATABASE_URL, and X OAuth/user tokens when you are ready to generate and publish for real.

Database Migrations

The backend uses Alembic for schema migrations.

cd apps/api
source .venv/bin/activate
alembic upgrade head

For Supabase, set DATABASE_URL in apps/api/.env using the Postgres connection string. Keep real credentials in .env; commit only .env.example.

Core Loops

• Research and strategy inputs: topics, positioning, audience, personality, emotion, forbidden phrases, CTA rules.
• Content generation: text-only posts, image-backed posts, threads, hooks, rewrites, scoring, and brand-fit review.
• Scheduling: manual calendar scheduling plus autonomous queue generation.
• Publishing: X OAuth-ready posting with media upload support and retryable publish logs.
• Operations: jobs, failures, retries, queue health, and content quality metrics.