Rub the lamp. Your wish is my command.
| Username | Role | Jobs | Last Active |
|---|
| Code | Created By | Expires | Used By |
|---|
v4 — March 16, 2026
Mobile-friendly hamburger menu
Live AI streaming now works end-to-end
In-browser app launch fixed
Profile tab with account info
Smarter job folder naming
Improved repair mechanism
v3 — March 15, 2026
Live AI streaming — watch code being written in real time
Home screen with example prompts
Genie branding & easter eggs
Email notifications on job completion
Browser push notifications
Provider presets (Groq, Together, Ollama…)
Launch generated web apps in-browser
Multi-turn: continue and improve past projects
v2 — March 12, 2026
Async jobs with cancel
File browser + ZIP download
WebDAV sync
Admin panel
Before you can run any jobs, go to Profile → OpenAI Configuration and paste your key (sk-…). Get one at platform.openai.com/api-keys. It's encrypted at rest and never shared with other users.
Type a plain-English description. Be specific — mention language, libraries, and expected output. Genie auto-detects script vs. project mode.
The AI generates a project plan, installs packages, writes code, and runs it — all inside an isolated sandbox. Watch the status update live.
Click Download ZIP for all generated files, configure WebDAV to auto-sync, use Continue to iterate on completed projects, or click Launch App to preview Flask/web apps live.
• Use gpt-4.1-mini for cheaper runs, gpt-4.1 for best quality.
• Sessions last 8 hours and auto-refresh.
• The sandbox has no display — GUI apps are built as Flask web apps automatically.
• Launched apps auto-stop after 15 minutes.