collage-broll

将约 5 秒口播文稿、观点句或抽象概念做成高级 editorial halftone paper-collage / 半调纸拼贴 B-roll。用户说"collage b-roll""纸拼贴 b-roll""半调拼贴""拼贴风格配画面""用这段文稿做拼贴动画"，或希望把一句文稿转成拼贴视觉隐喻时，必须使用此 skill。强制采用三阶段审批：先只提视觉隐喻，用户确认后用 siliconflow-img-gen 生成彩色拼贴静帧，静帧再次确认后用公共 aigc-video-gen 技能（i2v 首尾帧插值）组装动画。视频生成走百炼 happyhorse-1.1-i2v 候选链（沿链 fallback）或火山 Seedance（视 env 配置）。

Quality

94%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Quality

Content

88%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 strong, executable workflow skill: concrete commands and templates throughout, a rigorously gated three-phase process with both visual and technical validation, and a clean script bundle. The main gap is conciseness and some inline reference content that could be externalized.

Suggestions

Tighten the rationale prose under '强制审批协议' and '色彩规则' to directive statements; the two-gate justification and color-emotion mappings add tokens without adding executable value.

Consider moving the full Seedream and aigc-video-gen prompt templates plus the QA checklists into a references/ file (e.g. PROMPT-TEMPLATES.md, QA.md) and signaling them one level deep, which would lift progressive_disclosure to a clean 5.

Trim the duplicated model-fallback chain (it appears in both the description and Gate 3) to a single authoritative statement in the body.

Dimension	Reasoning	Score
Conciseness	Mostly efficient—every phase leads with executable commands, JSON schemas, or prompt templates and avoids explaining what Claude already knows. Not a 5 because some rationale prose (e.g. the two-gate justification, color-rule exposition) could be tightened further.	4 / 5
Actionability	Fully executable throughout—copy-paste ffmpeg tile/crop/audio-strip commands, aigc-video-gen i2v invocations with all flags, a gen-jobs.json schema, and complete Seedream/video prompt templates covering the common cases.	5 / 5
Workflow Clarity	Clear three-gate sequence with explicit validation checkpoints and feedback loops: Gate 1/2 user confirmation, 静帧 QA checklist, 视频 QA checklist, and a mandatory video-review technical gate (verdict=pass required) with regenerate-on-fail handling.	5 / 5
Progressive Disclosure	Well-organized into phased sections with a real one-level-deep script bundle (scripts/run_gate3.py, scripts/check_setup.sh) referenced and listed. Not a 5 because substantial reference-style content (color rules, QA checklists, prompt templates) lives inline rather than split into separate reference files.	4 / 5
	Total	18 / 20 Passed

Description

100%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 exemplary: it states concrete capabilities, lists natural user trigger phrases in the user's language, explicitly defines when to invoke, and carves out a distinctive niche. It uses appropriate third-person imperative voice throughout with no fluff.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions—'做成高级 editorial halftone paper-collage B-roll', '生成彩色拼贴静帧', 'i2v 首尾帧插值组装动画', '候选链 fallback'—covering the full capability comprehensively.	5 / 5
Completeness	Explicitly answers both what (turn a ~5s script into editorial paper-collage B-roll) and when ('用户说…必须使用此 skill') with concrete trigger phrases and a stated three-gate workflow.	5 / 5
Trigger Term Quality	Comprehensive natural Chinese trigger phrases users would actually say—'collage b-roll', '纸拼贴 b-roll', '半调拼贴', '拼贴风格配画面', '用这段文稿做拼贴动画'—with synonyms and intent variants.	5 / 5
Distinctiveness Conflict Risk	Clear niche—editorial halftone paper-collage B-roll with a specific three-gate i2v pipeline—highly distinct from generic video/image skills with minimal overlap risk.	5 / 5
	Total	20 / 20 Passed

Validation

87%

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 — 14 / 16 Passed

Validation for skill structure

Criteria	Description	Result
metadata_version	'metadata.version' is missing	Warning
metadata_field	'metadata' should map string keys to string values	Warning

	Total	14 / 16 Passed

Repository: TeamWiseFlow/xiaobei
Path: crews/content-producer/skills/collage-broll/SKILL.md
Commit: 72fd07a

Reviewed: 5 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.