video-cutter: standalone sentence-aware cutter

SkillMedia

Cut any video into sentence-aligned clips at silence-midpoint boundaries given approximate target timestamps. Works on YouTube downloads, Loom recordings, uploaded MP4s, recorded calls. Triggers on "cut this video at", "split into sections", "make clips from", "cut at boundaries".

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the video-cutter: standalone sentence-aware cutter skill

What this skill tells your AI

The instructions your AI receives, as published by naveedharri/benai-skills in shared-skills/video-cutter/SKILL.md and read by ahel’s review.

When triggered, invoke the video-cutter-agent sub-agent with:

  • video_path: path to the source video
  • word_timings_path: path to a word-timings JSON, or null (the agent will transcribe with faster-whisper if missing)
  • target_cut_points: approximate timestamps in seconds where the user wants cuts
  • output_dir: default Team/BenAI/Profiles/Milan/Daily/{today}/cuts/, or user-specified
  • section_titles: ordered names for the output clip filenames

If word timings are not provided, the agent will transcribe first. Pass null for word_timings_path to trigger this.

Profile boundary

Output writes to Profiles/Milan/ unless the user specifies otherwise. See [[yt-pipeline SPEC|SPEC.md]].

For the algorithm details, see [[algorithm]]. For QA verification, see [[qa-protocol]].

Signals

GitHub stars
64
Forks
29
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
video-cutter
Source
github.com/naveedharri/benai-skills