I cut a NVDA 2-hour keynote into 121 clips with one prompt - no hands

We cut every time Jensen Huang said "AI" in his CES keynote — turns out he said it 121 times. One prompt handled the entire workflow.

- Dive chained two MCPs together (yt-dlp + ffmpeg)

- Download video → parse subtitle timestamps → cut 121 clips → merge

- All processed locally

One prompt:

Task: Create a compilation video of every exact moment Jensen Huang says "AI".
Video source:

Instructions:

Download video in 720p + subtitles in JSON3 format (word-level timestamps)

Parse JSON3 to find every "AI" instance with precise start/end times

Use ffmpeg to cut clips (~50-100ms padding for natural sound)

Concatenate all clips chronologically

Output: Jensen_CES_AI.mp4

What is Dive?

Open-source MCP client that helps you install and use MCPs without touching the terminal. Local-first by design.

What is OAPHub?

Cloud MCP platform for tools you can't run locally — official MCPs (Figma, Sentry, Atlassian) and Pro MCPs for generative AI (Seedream, Kling, Flux).

We believe local-first. Dive handles what you can run yourself. OAPHub fills the gaps.

If you want to see how it runs:

📦 GitHub: github.com/OpenAgentPlatform/Dive