hermes-mcp/product/animaGen/README_local_render.md

Render the SquareMCP launch video locally

Files

• render_squaremcp_video.py renders the video
• squaremcp_shotlist.json drives scene timing and copy
• squaremcp_launch_captions.srt provides burned-in captions
• squaremcp_visual_prompts.md can be used to create optional scene background images

1. Install dependencies

python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt

2. Basic render

Place the script in the same folder as the shotlist and captions, then run:

python render_squaremcp_video.py \
  --shotlist squaremcp_shotlist.json \
  --captions squaremcp_launch_captions.srt \
  --output squaremcp_launch.mp4

3. Faster draft render

python render_squaremcp_video.py \
  --shotlist squaremcp_shotlist.json \
  --captions squaremcp_launch_captions.srt \
  --output squaremcp_launch_draft.mp4 \
  --draft

4. Add your own voiceover

python render_squaremcp_video.py \
  --shotlist squaremcp_shotlist.json \
  --captions squaremcp_launch_captions.srt \
  --voiceover founder_voiceover.wav \
  --output squaremcp_launch_with_vo.mp4

5. Optional richer visuals

If you generate background stills for each scene from the prompts, save them in an assets/ folder like this:

• assets/scene1.png
• assets/scene2.png
• ...
• assets/scene7.png

Then run:

python render_squaremcp_video.py \
  --shotlist squaremcp_shotlist.json \
  --captions squaremcp_launch_captions.srt \
  --assets-dir assets \
  --output squaremcp_launch_with_assets.mp4

Notes

• Default output is vertical 1080x1920
• --draft renders 720x1280 and is much faster
• The renderer creates a clean product-style motion graphic even without scene images
• If you already have a polished voiceover, attach it with --voiceover