TimoteoAltobrando/ai-video-skill
ai-video
UnverifiedMake and edit videos with Claude Opus 5.5, end to end, from a blank page or from the user's own footage — product and launch films, ads, talking-head and founder videos, interviews, explainers, tutorials and screen demos, event and trip montages, highlight reels, podcast clips, Reels/TikTok/Shorts, animated logos and intros, music and lyric videos, and changes to a finished video (shorter cut, vertical version, subtitles, translation, new music). Remotion is the backbone; Codex (image_gen) makes images, Seedance via OpenRouter animates them, ElevenLabs makes music and sound effects, ffmpeg and whisper handle footage and speech. It starts by asking what kind of video it is and everything that kind needs (platform, format, voice, music, rights, who must not appear, budget), then goes through direction, a script or paper edit approved piece by piece, production, revisions and delivery for each platform. Use it whenever someone wants to make, cut, edit, shorten, reformat, subtitle or improve a video, even if they do not mention any of these tools.
scripts/export.sh
View on GitHub#!/bin/sh
# Export a Remotion composition with the audio in sync and mastered, plus a light 720p copy.
#
# sh export.sh <entry.ts> <CompositionId> <output.mp4> [more remotion options, e.g. --frames=0-47]
# FIXED_GAIN_DB=4.5 sh export.sh ... # reuse the master gain of a previous export instead of re-normalising
#
# Why it works this way:
# - Remotion 4's default mp4 (AAC) plays about 43 ms after the picture: here the picture is rendered without audio,
# the audio as WAV (exact), and ffmpeg muxes the two with the right priming (zero delay);
# - the audio is brought to -14 LUFS integrated (the web and social standard) with a limiter at -1.5 dBFS, so the
# true peak stays around -1 dBTP;
# - when only the music (or only the effects) changes, re-normalising to -14 LUFS raises everything else: the voice
# and the effects get louder and the change you asked for shrinks. Pass FIXED_GAIN_DB with the gain printed by the
# previous export to keep the voice exactly where it was;
# - Remotion sums all the audio tracks into 16-bit PCM: if loud moments go over full scale they are clipped before
# the limiter can help. The script counts clipped samples and warns: lower the volumes in the composition (e.g. all
# tracks x0.75) and raise the gain here instead;
# - --timeout=120000: with video clips a frame is sometimes late and the render would stop.
# Outputs: <output.mp4> (high quality, AAC 320k) and <output>-720p.mp4 (for the phone, about 5-6 MB per 50 s).
set -e
ENTRY="$1"; ID="$2"; OUT="$3"; shift 3
TMP="$(mktemp -d)"
npx remotion render "$ENTRY" "$ID" "$TMP/video.mp4" --crf=18 --muted --timeout=120000 "$@"
npx remotion render "$ENTRY" "$ID" "$TMP/audio.wav" --codec=wav --timeout=120000 "$@"
python3 - "$TMP/audio.wav" <<'EOF'
import sys, wave, array
w = wave.open(sys.argv[1])
if w.getsampwidth() == 2:
a = array.array("h", w.readframes(w.getnframes()))
n = sum(1 for v in a if v >= 32766 or v <= -32767)
if n:
print(f"WARNING: {n} clipped samples in the rendered audio: lower the volumes in the composition and re-export")
EOF
measure() { ffmpeg -hide_banner -nostats -i "$1" -af ebur128=peak=true -f null - 2>&1 | awk '/Summary:/{s=1} s&&/ I:/{i=$2} s&&/Peak:/{p=$2} END{print i, p}'; }
set -- $(measure "$TMP/audio.wav")
if [ -n "$FIXED_GAIN_DB" ]; then G="$FIXED_GAIN_DB"; else G=$(python3 -c "print(round(-14.0 - ($1), 2))"); fi
echo "audio before: $1 LUFS, true peak $2 dBTP; gain $G dB (reuse it with FIXED_GAIN_DB=$G)"
ffmpeg -v error -y -i "$TMP/audio.wav" -af "volume=${G}dB,alimiter=limit=0.84:attack=1:release=60:level=false" -c:a pcm_s24le "$TMP/audio-m.wav"
set -- $(measure "$TMP/audio-m.wav")
echo "audio after: $1 LUFS, true peak $2 dBTP"
ffmpeg -v error -y -i "$TMP/video.mp4" -i "$TMP/audio-m.wav" -map 0:v -map 1:a -c:v copy -c:a aac -b:a 320k -shortest -movflags +faststart "$OUT"
LIGHT="${OUT%.mp4}-720p.mp4"
ffmpeg -v error -y -i "$OUT" -vf scale=1280:-2 -c:v libx264 -preset slow -crf 25 -pix_fmt yuv420p -c:a aac -b:a 192k -movflags +faststart "$LIGHT"
rm -rf "$TMP"
echo "-> $OUT"
echo "-> $LIGHT"
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |