render-song-mv

Assemble a song-driven music-video ad from a config — a generated sung track carries the whole narration across N tableaux (one keyframe -> one i2v clip per lyric beat) with NO separate voiceover, captions synced to the song's OWN word timings (script-window, never Whisper) and the hook word landing on the chorus drop, closed on a PIL brand end card. This is the FREE deterministic assembly stage (clip cut-to-timeline + captions + end card + FFmpeg composite); the song, keyframes, and clips come from create-music-elevenlabs / create-image-fal / create-video-fal. Use for the song-driven-music-video format.

Quality

61%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

tessl review fix ./skills/ads/capabilities/render-song-mv/SKILL.md

Quality

Content

46%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

The body is a well-structured overview/contract with good bundle-file navigation, but it is descriptive rather than executable and lacks validation checkpoints for a destructive/batch assembly. Adding concrete FFmpeg/PIL commands and verify steps would lift actionability and workflow clarity.

Suggestions

Add an executable Run sequence with the actual FFmpeg composite command(s) (cut/concat, ASS burn, end-card overlay, loudnorm, mux to 1080×1920 h264+aac) instead of describing them in prose.

Insert validation checkpoints: verify words.json clip/beat count matches tableaux, verify output is 1080×1920, verify loudnorm hit −14 LUFS, before declaring the master done.

Convert the bundle mentions to markdown links (e.g., '[PIPELINE.md](scripts/PIPELINE.md)') so navigation is explicit rather than implied.

Dimension	Reasoning	Score
Conciseness	Mostly efficient prose that assumes Claude knows FFmpeg/PIL/i2v, but the 'no separate VO / sung song carries narration' point is restated across the intro, Run, and Contract sections and could be tightened.	3 / 5
Actionability	The body states targets ('burn the caption ASS', 'loudnorm to −14 LUFS', 'chunk words.json ~3 words') but provides no executable FFmpeg commands, PIL code, or script invocations — it describes the assembly rather than instructing with runnable steps.	2 / 5
Workflow Clarity	A rough conceptual sequence exists (song → timeline → captions → hook → cuts → end card → FFmpeg) as a contract list, but there is no numbered run procedure and no validation checkpoints; per the rubric, a batch/destructive assembly without validation is capped at 3.	3 / 5
Progressive Disclosure	Well-organized sections with three real, clearly signaled one-level-deep references to existing bundle files (config.example.json, PIPELINE.md, README.md); references are prose mentions rather than markdown links, keeping it just below the top anchor.	4 / 5
	Total	12 / 20 Passed

Description

76%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

The description is specific and clearly distinct, naming concrete assembly actions and an explicit use-format trigger. Its main gap is trigger-term breadth: it relies on a single jargon-leaning phrase rather than the natural synonyms a user would say.

Suggestions

Add natural trigger synonyms to the 'Use for...' clause (e.g., 'Use when making a song-driven music video, a lyric-timed video ad, or a sung-narration spot').

Trim the dense parenthetical jargon ('one keyframe -> one i2v clip per lyric beat') to keep the description scannable for trigger matching.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete assembly actions — 'clip cut-to-timeline', 'captions', 'end card', 'FFmpeg composite', 'captions synced to word timings', 'hook word landing on chorus drop' — giving comprehensive coverage of what the stage does.	5 / 5
Completeness	A clear 'what' (assembles a song-driven music-video ad from a config) is paired with an explicit 'Use for the song-driven-music-video format' clause; the 'when' is present but terse and could name more trigger phrases.	4 / 5
Trigger Term Quality	Has the natural trigger 'song-driven-music-video format' but coverage is thin and leans on jargon ('tableaux', 'i2v clip'); common synonyms a user would say ('music video', 'song video ad') are missing.	3 / 5
Distinctiveness Conflict Risk	A sharply defined niche — sung-track-as-narration, no voiceover, cited competing capabilities — with a distinctive trigger leaves minimal overlap risk with other skills.	5 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: gooseworks-ai/goose-skills
Path: skills/ads/capabilities/render-song-mv/SKILL.md
Commit: 2523432

Reviewed: 7 days ago

Table of Contents

Discovery Implementation Validation

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.