VEXORA-Code-AI

VEXORA portfolio demo

Demo account setup

Use a fresh local development database and create a demo-only account through the UI. Do not use production credentials or real customer code in a portfolio recording. The workspace starts with sample Python and JavaScript files after the first project is created.

For a polished recording, configure a server-side AI key and build the isolated execution image first. Without those optional services, the editor, project persistence, sign-in, and shell-free output panel still work; AI shows a configuration error and execution fails closed.

90-second walkthrough

  1. Sign in and show the private project/file tree and editor tabs.
  2. Open src/main.py, make a small edit, then refresh or switch files to show saved project state.
  3. Select the Python language and run the sample; show the bounded output panel.
  4. Ask the assistant to explain the greeting function, then demonstrate a code review request.
  5. Point out the separate output, terminal, and problems panels, and the responsive workspace layout.
  6. Close on the implementation boundaries: server-side provider credentials, owner-scoped MySQL records, CSRF-protected sessions, and a separate restricted runner.

Keep any AI prompt generic and non-sensitive. Do not expose .env, account credentials, provider keys, or database contents in screenshots or screen recordings.

Screenshot checklist

Screenshots in docs/screenshots/ are captured from the local app at two viewport sizes and linked from the root README. Add these signed-in workspace views after preparing a disposable local demo account:

Capture from a demo-only account after hiding personal information and browser password managers. For automated capture, preserve a demo DB snapshot and never reuse its credentials in production.