render-vignette

Assemble a short-form 'vignette' ad from clean product cutouts composited over a kinetic background video — birefnet cutout, then a cold-open text card + product carousel + annotated specimen-sheet end card, plus loudnorm + separate-pass music mux. FREE assembly (PIL + rsvg + FFmpeg); the recipe supplies the config and gates the paid BG/cutout/music calls to their own capabilities. Use for the vignette format.

Quality

72%

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-vignette/SKILL.md

Quality

Content

68%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 concise, concrete, and well-structured with real bundled scripts and exact processing parameters. Its chief gap is the absence of explicit validation/feedback-loop checkpoints in the Run sequence for a batch operation, which caps workflow clarity.

Suggestions

Add explicit validation checkpoints to the Run sequence, e.g. after step 1 verify the cutout manifest alpha quality, and after step 3 probe each variant's audio bitrate to confirm the mux shipped real audio before proceeding.

Show the actual run invocations (e.g. `python3 scripts/strip_product_backgrounds.py`) in the Run steps so the sequence is copy-paste executable, not just script names.

Tighten the Gaps/routing-notes section by collapsing the repeated prod-routing explanation into a one-line pointer per gap.

Dimension	Reasoning	Score
Conciseness	Efficient and assumes Claude's competence — no padding explaining FFmpeg/PIL/birefnet — with specific filter values that earn their place; the Gaps/routing notes prose is the one section that could be trimmed.	4 / 5
Actionability	Names the four real, executable scripts and gives concrete FFmpeg/audio parameters (loudnorm I=-18:TP=-2:LRA=9, h264 crf20 yuv420p +faststart, y=(H-h)/2, saturation=0.50), though the Run steps name scripts rather than giving copy-paste run invocations.	4 / 5
Workflow Clarity	The four steps are clearly sequenced, but this is a batch operation (6 variants) and the body surfaces no explicit validate→fix→retry checkpoints in the Run flow, so workflow clarity is capped at 3 per the batch-operation rule.	3 / 5
Progressive Disclosure	Well-organized into Run/Contract/Gaps with the bundled scripts clearly named and described as one-level-deep executable material; no nested references, though the scripts are referenced as run steps rather than via explicit 'see X for details' navigation.	4 / 5
	Total	15 / 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, comprehensive, and clearly niched, with an explicit 'what' and 'when'. Its main weakness is jargon-heavy trigger language and a narrow, single-clause use-condition that misses natural user phrasings and synonyms.

Suggestions

Broaden the trigger clause beyond 'Use for the vignette format' to natural user phrasings, e.g. 'Use when building short-form product ads, vertical 9:16 vignettes, or muted loopable product videos.'

Surface a few user-facing synonyms (short-form ad, vertical product video, 9:16 ad) alongside the internal jargon so the description triggers on natural language.

Consider stating the format constraints (sub-12s, 9:16, music-led/no-VO) in the description to further distinguish from generic video compositing skills.

Dimension	Reasoning	Score
Specificity	Names the domain and a comprehensive set of concrete actions — 'birefnet cutout', 'cold-open text card + product carousel + annotated specimen-sheet end card', 'loudnorm + separate-pass music mux', 'PIL + rsvg + FFmpeg' — covering cutout, overlay, composite, and audio stages.	5 / 5
Completeness	Clearly states what (assemble a vignette ad from cutouts over a BG video with the listed cards and audio) and gives an explicit 'Use for the vignette format' trigger, but the when-clause is a single terse phrase rather than varied concrete trigger phrases.	4 / 5
Trigger Term Quality	Contains the natural term 'vignette ad'/'vignette format' but is dominated by internal jargon (birefnet, loudnorm, separate-pass mux, specimen-sheet) users would not naturally say, and lacks common synonyms/variations.	3 / 5
Distinctiveness Conflict Risk	Carves a clear niche (short-form 'vignette' ad with a named V-CARD structure) with a distinct trigger ('vignette format'), giving minimal overlap with generic video 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-vignette/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.