Content Studio is an end-to-end platform for WeChat official account operations, covering subscription monitoring, article ingestion, AI analysis, AI writing, and rewriting workflows.
- Phone + password registration and login
- Per-user data isolation (subscriptions, articles, tags, tasks, AI profiles)
- OpenAI-compatible model configuration (
base_url,api_key,model) - One-click
Analyze/Create/Rewriteper article - Creator workbench overview (assets, draft count, plan quota)
- Tiered plans for
Free/Pro/Premiumusers - Local draftbox + optional one-click sync to WeChat draftbox
- Draft publish queue (auto enqueue on failure + retry)
- Admin plan console (tier/quota adjustments)
- Unified web UI for subscriptions, tasks, tags, and system settings
- Containerized deployment with PostgreSQL support
pip install -r requirements.txt
cp config.example.yaml config.yaml
python init_sys.py
python main.py -job True -init FalseOpen: http://127.0.0.1:8001
cd compose
docker compose -f docker-compose-postgresql.yaml up -d --buildOpen: http://127.0.0.1:8001
Default admin account:
- Username:
admin - Password:
admin@123
In config.yaml:
db: database DSN (PostgreSQL recommended)secret: JWT signing secretserver.web_name: web app titletoken_expire_minutes: auth token lifetimerss.*: RSS output behavior
PostgreSQL example:
postgresql://rss_user:pass123456@postgres:5432/we_mp_rss
Go to AI Studio after login and configure:
Base URLModelAPI Key
Then run one-click actions on each article:
- Analyze
- Create
- Rewrite
After generation:
- Save to local draftbox
- Optionally sync to WeChat draftbox (requires QR authorization and proper plan tier)
- Optional Jimeng image generation (falls back to prompts when AK/SK are missing)
- Startup and usage guide:
docs/guides/启动与使用指南.md - History cleanup script:
script/clean_history.sh
This delivery includes:
- unified product UI and layout
- phone-based user registration/login
- tenant-level data isolation
- AI content studio for article-level generation
- PostgreSQL containerized deployment
- Playwright E2E flow verification