ugc-fixloop

the UGC fix-loop toolkit — surgically re-render a bad window/beat of a single-take UGC master (stitch_replacement.py, pure FFmpeg) and GPT cross-model review a Seedance prompt before render (vet_seedance_prompt.py, routed through the openai-proxy). Fetch it into a one-shot UGC recipe so both scripts resolve on any machine and the vet call bills the Ads agent.

Quality

76%

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/ugc-fixloop/SKILL.md

Quality

Content

93%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 an exemplar of lean, actionable skill content: executable commands for both bundled scripts, a clear safety invariant, and well-organized structure that delegates detail to the one-level-deep script docstrings.

Dimension	Reasoning	Score
Conciseness	The body is lean and assumes Claude's competence: it skips what FFmpeg/Seedance are, gives only the runnable flags and the one safety invariant, and every line earns its place with no concept padding.	5 / 5
Actionability	It provides copy-paste-ready command invocations for both scripts with all required and optional flags enumerated, plus concrete examples for beat-based and explicit-window selection covering the common cases.	5 / 5
Workflow Clarity	The two scripts' usage is clearly sequenced with args and a duration-drift validation warning, but the overall fix-loop flow (which script runs first, how vet feeds the render) is implied rather than an explicit numbered sequence with checkpoints.	4 / 5
Progressive Disclosure	The SKILL.md is a compact overview that points one level deep to the two bundled scripts in ./scripts/ (both verified present), with each script's docstring carrying the detail; structure is well-organized and navigable for a sub-50-line skill.	5 / 5
	Total	19 / 20 Passed

Description

58%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 distinctive with concrete tool actions, but it omits an explicit 'Use when...' trigger clause, capping completeness, and its trigger terms are domain-jargon-heavy rather than natural user phrasing.

Suggestions

Add an explicit 'Use when...' clause naming the trigger situations (e.g. 'Use when a rendered UGC master has one bad beat that needs a surgical re-render, or before spending a Seedance render to cross-check the prompt').

Soften jargon with natural synonyms a user would say — 'fix a bad clip', 'swap out one beat', 'second-opinion review of a video prompt' — alongside Seedance/UGC terms.

Lead with the core verb actions before the tool filenames so the capability reads as actions rather than a file manifest.

Dimension	Reasoning	Score
Specificity	Names the domain (UGC fix-loop) and several concrete actions (surgically re-render a window/beat, GPT cross-model review a Seedance prompt, fetch into a one-shot recipe), but the actions lean slightly tool-naming rather than a tight verb list, leaving minor coverage gaps.	4 / 5
Completeness	It clearly states what the toolkit does, but there is no explicit 'Use when...' trigger clause telling Claude when to invoke it, which caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	It references domain terms (UGC, Seedance prompt, stitch replacement, fix-loop) but lacks the natural conversational phrases and synonyms a user would actually say (e.g. 'fix a bad clip', 're-render a beat', 'review a video prompt'), so common variations are missing.	3 / 5
Distinctiveness Conflict Risk	The niche is quite specific (single-take UGC Seedance masters, surgical beat swap, cross-model prompt vetting via a named proxy) and unlikely to fire for unrelated skills, with only minor overlap risk against general video-editing skills.	4 / 5
	Total	14 / 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/ugc-fixloop/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.