gooseworks-ai/goose-skills / skills/video/render-cinematic-music-video
render-cinematic-music-video
VerifiedAssemble a cinematic live-action-style music-video ad from a config — an original sung anthem carries the whole narrative while N 35mm-film-look i2v clips are each cut to their lyric window and hard-concatenated on the beat as a 3-act arc, the anthem muxed at loudnorm I=-14, cinematic lower-third serif captions built from the song's OWN word timings (never Whisper) with the hook line landing on the chorus drop, and closed on a brand end card composited from the real asset — never AI-rendered text. This is the FREE deterministic assembly stage (cut-to-window + hard concat + anthem mux + captions + end card); the anthem, keyframes, and clips come from create-music-elevenlabs / create-image-gpt-image-fal / create-video-fal. Use for the cinematic-music-video format.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 5.3 KB |
| scripts/PIPELINE.md | markdown | 10.0 KB |
| scripts/README.md | markdown | 2.9 KB |
| scripts/config.example.json | config | 22.9 KB |
| skill.meta.json | config | 211 B |
| tests/smoke-test.md | markdown | 1.2 KB |
scripts/README.md
View on GitHub# render-cinematic-music-video scripts — the FREE assembly
`render-cinematic-music-video` is the **deterministic, free assembly stage** of the cinematic
music-video format. The paid stages (the sung anthem, the N film-look keyframes, the N Kling i2v
clips) are separate capabilities — `create-music-elevenlabs`, `create-image-gpt-image-fal`,
`create-video-fal`. This capability spends nothing: it takes the delivered anthem + `words.json`
+ one clip per beat + the brand end-card asset and stitches the finished master. Re-cuts (new
caption chunking, a swapped end card, re-timed windows) reuse the existing anthem / keyframes /
clips and spend nothing.
`config.example.json` is the worked example (Sundrift "Roll On", ~28s 1080×1920).
Its song style, vocalist, cast, setting and story arc are the demo's answers to the format's
choices — a new run gets them from the user; copy the structure, not the values.
`PIPELINE.md` maps every config block to its source step. This README documents the FREE assembly
pieces that `render-cinematic-music-video` owns.
## 1. Lower-third captions — from the anthem's OWN word timings (NOT Whisper)
Captions come from `audio/words.json` — derived from the music model's `words_timestamps` (ms →
s) — never from Whisper (Whisper on sung audio returns "🎵 Music Playing 🎵"). The builder chunks
~4 words at lyric boundaries, times each event, and burns cinematic lower-third serif captions
(low on the frame, serif font) with the `stitch-videos-ffmpeg` skill's `montage.py captions`; the
chorus hook line and accent words go on in a second pass in the accent colour. The hero/hook chunk is timed so the load-bearing line lands on the chorus drop.
## 2. Per-beat clip assembly to the anthem timeline
The anthem sets the timeline. Each tableau boundary was snapped to a lyric-phrase edge in the
returned word timings, so assembly cuts each per-beat clip to its lyric window
(`scale=1080:1920 increase` + `crop`, `fps=30`, `tpad clone` to the window — pair `tpad` with an
explicit `-t` clamp) and hard-concats the clips **on the beat** (one optional match-cut into the
hero reveal). No dissolves. The single look pack + continuity anchor make N beats read as shot by
one photographer.
## 3. End card — from the real brand asset, no AI text
The brand lockup is composited (PIL/ffmpeg) from the brand's REAL asset, or authored as a
designed keyframe base, overlaid on the final window. The brand text is **never** AI-rendered as
free diffusion text — a diffusion model garbles a wordmark.
## 4. FFmpeg composite
FFmpeg stitches the master: cut each clip to its lyric window, hard-concat on the beat →
`master-silent.mp4`; mux the anthem (`atrim`, `afade` in 0.15s / out 0.6s + `loudnorm I=-14`) →
`master-no-captions.mp4`; burn the cinematic lower-third captions with `montage.py captions` →
`master-final.mp4`. The sung anthem IS the bed — no separate VO. Output is a 1080×1920 h264 + aac
master. Deterministic, no paid calls, no keys.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack." | Stars: 198k | Updated | Claude CodeCodex | Verified | Install |
| ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt. | Stars: 196k | Updated | Claude CodeCodex | Verified | Install |