agentsclimarketplace

Run3 extract keyframes from video

Skill cxcscmu/SkillLearnBench/skills/b3-teacher-feedback-gemini-3-flash-preview/video-object-counting/run3_extract_keyframes_from_video

[COLM'26] SkillLearnBench is the first benchmark for evaluating continual learning methods that automatically generate agent skills.

Install
npx -y skills add cxcscmu/SkillLearnBench --skill run3_extract_keyframes_from_video

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

What its author says it does

Copied from the file, not written here

Converts a video file into a series of keyframe images using FFmpeg with specific frame selection filters.

SKILL.md

0.8 KB, as published. Nobody here has run it

To extract high-quality keyframes from the video, use FFmpeg with the I-frame selection filter. This ensures only the primary frames are captured for analysis.

Command:

ffmpeg -i /root/super-mario.mp4 -vf "select='eq(pict_type\,I)'" -vsync vfr /root/keyframes_%03d.png

Parameters:

  • -vf "select='eq(pict_type\,I)'": Filters the video to select only Intra-coded frames (keyframes).
  • -vsync vfr: Ensures variable frame rate is handled correctly so no frames are duplicated or dropped.
  • /root/keyframes_%03d.png: The output pattern to save frames in chronological order (e.g., keyframes_001.png, keyframes_002.png).

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.