# Q-Star AI 2-Minute Demo Shot List

## Demo goal

Prove that Q-Star AI is an operating multi-agent AI creation infrastructure for video, apps, and automated creative workflows.

The demo should not feel like a slide deck. It should show a real prompt moving through planning, model routing, task execution, media/app generation, storage, and final output.

## Timing plan

| Time | Shot | Screen to show | Voiceover |
| --- | --- | --- | --- |
| 0:00-0:10 | Opening | Q-Star AI homepage at https://ai.q-star.ink/ | Q-Star AI is building a multi-agent AI creation infrastructure for video, apps, and automated creative workflows. |
| 0:10-0:20 | Use case | Type or show prompt: "Create a short cinematic AI trailer and companion mobile app concept." | A user starts with one creative request. |
| 0:20-0:35 | Agent planning | AgentOS / OpenClaw task breakdown, workflow graph, or logs | The agent layer decomposes the request into story, assets, scenes, audio, UI, code, rendering, and review tasks. |
| 0:35-0:50 | Model routing | Model-routing logs or provider selection screen | The router assigns tasks to text, code, image, audio, video, and app-generation models. |
| 0:50-1:05 | Task queue | Queue, worker logs, render jobs, or async execution screen | Async workers process generation and rendering tasks with retries and status tracking. |
| 1:05-1:20 | Media generation | Image/video/audio generation output folder, render progress, or sample video clips | The pipeline generates media assets, renders video, and stores outputs. |
| 1:20-1:35 | App generation | APK Generator prompt, build log, output APK, or generated app UI | The same infrastructure can also produce app prototypes and build artifacts. |
| 1:35-1:50 | Final output | Generated video, generated app/prototype, and completed workflow state | The platform returns generated outputs and a workflow audit trail. |
| 1:50-2:00 | Google Cloud close | Evidence page, cloud architecture, and usage projection | Google Cloud credits would help us scale Gemini API, Cloud Run, Cloud Storage, Pub/Sub, Monitoring, and GPU workloads. |

## Required screen captures

- Q-Star AI homepage
- Evidence page
- Prompt input
- Agent planning or workflow graph
- Model routing logs
- Queue or worker progress
- Generated video output
- Generated app or APK output
- Storage/output folder
- Monitoring, server logs, or Docker/service view
- Cloud architecture diagram

## Minimum acceptable demo

If the full product flow is not ready for recording, use this minimum proof path:

1. Show homepage and evidence page.
2. Show a real terminal or dashboard prompt.
3. Show agent/task logs.
4. Show generated media output from the deployed `/videos` directory.
5. Show running backend services or Docker containers.
6. Show the Google Cloud migration plan.

## Recording notes

- Use English UI labels or English subtitles.
- Avoid explaining unrelated small tools.
- Keep the camera on real running screens.
- Do not show secrets, API keys, environment variables, private customer data, or billing details.
- Use 1080p or higher resolution.
- Keep the final video under 2 minutes 15 seconds.