raw-video-processing — Claude Code skill: setup and usage
Cleans up raw screen recordings by cutting silent pauses and speeding up the footage. Useful before publishing demos or tutorials.
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a claude-codeWhat raw-video-processing does
The skill streamlines rough screencast editing using two Python scripts managed by uv. It first triggers remove_silence.py. That script cuts out dead air longer than 0.5 seconds without re-encoding the stream. The aggressive -20dB threshold treats keyboard typing and ambient noise as silence, keeping pauses out of the finished recording.
Once silent sections are gone, speed_video.py applies a 1.2x playback rate and re-encodes the file through FFmpeg. Without ffmpeg and uv pre-installed on the machine, processing cannot run. Order matters here. Speeding up the clip beforehand changes audio pitches, which confuses the silence detector and cuts spoken parts.
What it helps with
- cut silent gaps and keyboard clatter from raw screencasts
- speed up edited screen capture video to 1.2x
- preview cut timestamps using
--dry-runbefore rendering - trim long software walkthroughs without manually slicing clips
Example prompts for the agent
- Remove dead air from
demo.mp4and boost playback speed to 1.2x - Run
remove_silence.pyonscreencast.mp4with--dry-runto inspect timestamps - Strip pauses and typing noise from
tutorial.mp4using a-20dBthreshold
How to install raw-video-processing
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a claude-code2Manually: put the files in place
~/.claude/skills/raw-video-processing/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/raw-video-processing/
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a codex2Manually: put the files in place
.agents/skills/raw-video-processing/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a cursor2Manually: put the files in place
.cursor/skills/raw-video-processing/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install zc277584121/marketing-skills raw-video-processing2Manually: put the files in place
.github/skills/raw-video-processing/SKILL.md1Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a gemini-cli2Platform CLI
gemini skills install https://github.com/zc277584121/marketing-skills --path skills/raw-video-processingManually: put the files in place
.devin/skills/raw-video-processing/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add zc277584121/marketing-skills --skill raw-video-processing -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about raw-video-processing
How do I install raw-video-processing in Claude Code?
Run npx skills add zc277584121/marketing-skills --skill raw-video-processing -a claude-code in your terminal.
Why must silence removal run before speeding up the clip?
Silence detection relies on intact frequencies from the original audio track. Speeding up the clip first warps the audio, causing the detector to misidentify speech or overlook quiet pauses.
Which system dependencies are required?
Both ffmpeg and uv must be present in your system environment alongside Python 3.12.