ワンクリックで
podguy-clip-cutter
// Use when the user wants to cut/export likely clippable podcast moments for TikTok, Reels, YouTube Shorts, trailers, or social review. Turn pi-selected clip candidates into media files with scripts/cut_clips.py.
// Use when the user wants to cut/export likely clippable podcast moments for TikTok, Reels, YouTube Shorts, trailers, or social review. Turn pi-selected clip candidates into media files with scripts/cut_clips.py.
| name | podguy-clip-cutter |
| description | Use when the user wants to cut/export likely clippable podcast moments for TikTok, Reels, YouTube Shorts, trailers, or social review. Turn pi-selected clip candidates into media files with scripts/cut_clips.py. |
| compatibility | Requires ffmpeg for actual cutting. Works from repo root after transcript prep and clip candidate selection. |
Use this skill when the user asks to cut, export, render, or prepare short-form/social clips from a podcast or video podcast.
This skill complements podguy-post-production:
scripts/cut_clips.py does deterministic ffmpeg cuttingdist/analysis/When a real video-podcast fixture is useful, run:
scripts/download_sample_media.sh
This downloads a local excerpt from Cordkillers 572 under dist/test-fixtures/open-license/cordkillers-572/ with attribution. The default 00:08:00-00:11:50 excerpt includes a three-person podcast layout, lower thirds, chat/sidebar graphics, a Patreon bumper, and an outro/interstitial card.
Identify:
ep006.dist/analysis/<slug>/clips.md.txt, .csv, or .json with start/end rangesIf the user has not selected clip candidates yet, run the normal clip workflow first and write dist/analysis/<slug>/clips.md.
The cutter accepts ranges in Markdown/plain text, CSV, or JSON.
Markdown example:
## Why this market flipped
- 00:12:14 - 00:13:08 — Strong contrarian claim with a clean payoff.
CSV example:
title,start,end
Why this market flipped,00:12:14,00:13:08
JSON example:
{
"clips": [{ "title": "Why this market flipped", "start": "00:12:14", "end": "00:13:08" }]
}
Original-aspect review cuts:
uv run python scripts/cut_clips.py \
<input-media> \
dist/analysis/<slug>/clips.md \
dist/analysis/<slug>/clips/cuts
Vertical review exports for TikTok/Reels/YouTube Shorts:
uv run python scripts/cut_clips.py \
<input-media> \
dist/analysis/<slug>/clips.md \
dist/analysis/<slug>/clips/shorts \
--aspect vertical \
--pad-start 1 \
--pad-end 1
Useful options:
--limit N cuts only the first N parsed clips--pad-start SECONDS and --pad-end SECONDS add review handles--aspect source|vertical|square|horizontal controls simple framing--copy is faster for original-aspect cuts but may be less frame-accurate--dry-run writes/prints commands without producing mediaRecommended paths:
dist/analysis/<slug>/clips.md — selected candidatesdist/analysis/<slug>/clips/cuts/ — original-aspect review exportsdist/analysis/<slug>/clips/shorts/ — vertical review exportsdist/analysis/<slug>/clips/*/manifest.json — machine-readable export manifestFor social/short-form exports, prefer candidates that:
Avoid exporting moments that: