video-creation-pro

电商运营与自媒体创作者在为美妆、家居、智能硬件等商品制作短视频时，请使用此技能。10大AI智能体协同，从卖点提炼、分镜生图到音效字幕，一键调用COZE API自动合成高质量商品视频，彻底告别本地代码依赖，轻松实现从创作、质检到数据迭代的全流程闭环。

Quality

64%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/video-creation-pro/video-creation-pro/SKILL.md

Quality

Content

58%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 content delivers a well-sequenced, validation-rich 10-agent workflow with good progressive-disclosure structure pointing to real reference files. Its weak points are heavy redundant repetition of the same prohibition (conciseness) and the absence of any executable COZE API call example or reference to the bundled scripts (actionability).

Suggestions

Consolidate the repeated '禁止本地依赖包/COZE API适配' statements into a single 前置准备/核心原则 block and reference it once per step instead of restating it 15+ times to reduce padding.

Add at least one executable COZE API call example (request payload or curl/Python snippet) in 步骤8, and either reference or explicitly justify omitting video_composer.py and image_quality_checker.py from scripts/.

Link or account for the unreferenced bundle files (development_guide.md, mask_video_guide.md, quality_standards.md, video_specs.md) in 资源索引 so all bundle materials are discoverable.

Dimension	Reasoning	Score
Conciseness	The ~340-line body is noticeably verbose, repeating 'COZE API适配', '无本地依赖包', and '无本地合成' across nearly every step (15+ times) with padded reinforcement of the same prohibition, matching the anchor for 2 rather than the mostly-efficient 3.	2 / 5
Actionability	Concrete spec is provided (input requirements, output standards, and API parameters like 1080P/25fps/MP4/transition 0.3-0.5s/retry ≤3/timeout 10s), but there is no executable COZE API call — no curl, request/response payload, or code sample — and the existing scripts (video_composer.py, image_quality_checker.py) are never referenced, matching the 'description instead of executable code' anchor for 3.	3 / 5
Workflow Clarity	The 10-step pipeline has an explicit architecture diagram, per-step trigger conditions, collaboration rules, and strong validation checkpoints (质检智能体, 素材预检, 重试机制, 驳回重做, '100%解决后方可进入下一环节') with feedback loops, matching the anchor for 5; the batch-operation cap does not apply because validation is present.	5 / 5
Progressive Disclosure	The body is an overview with a clearly signaled, one-level-deep 资源索引 pointing to 6 real reference files (all verified present), but 4 bundle files (development_guide, mask_video_guide, quality_standards, video_specs) and the entire scripts/ directory are unreferenced while the body says '必要脚本：无', leaving minor organization gaps that fit anchor 4 rather than 5.	4 / 5
	Total	14 / 20 Passed

Description

70%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 clearly conveys both capability and trigger context for a specific niche, with strong concrete actions and good keyword coverage. Its main weaknesses are second-person voice (which costs specificity) and a 'when' clause that is audience-based rather than crisp trigger phrases.

Suggestions

Rewrite in third person (e.g., 'Creates product short videos...' instead of '请使用此技能') to avoid the specificity penalty for second-person voice.

Add explicit natural trigger phrases to the 'when' clause, such as 'Use when the user wants to generate product promo videos, short-video ads, or 卖点 videos for 美妆/家居/智能硬件 goods.'

Include file-format and synonym triggers (e.g., .mp4, 商品短视频, 带货视频, 种草视频) to lift trigger term coverage toward 5.

Dimension	Reasoning	Score
Specificity	The description names the domain and multiple concrete actions (卖点提炼, 分镜生图, 音效字幕, COZE API合成, 质检, 数据迭代), which would anchor at 4, but the second-person phrasing '请使用此技能' triggers the mandated -1 specificity penalty, bringing it to 3.	3 / 5
Completeness	It states both what the skill does (10 agents synthesizing product videos via COZE API through the full pipeline) and when to use it (when e-commerce/self-media creators make short videos for products), though the 'when' is audience-framed and could be more trigger-phrase explicit, matching anchor 4.	4 / 5
Trigger Term Quality	Good coverage of natural terms a user might say (商品短视频, 美妆/家居/智能硬件, 卖点, 分镜, 字幕, 音效), with only a few common variations or file extensions missing, matching the anchor for 4.	4 / 5
Distinctiveness Conflict Risk	The niche (COZE-API product short-video synthesis with 10 specialized agents) is mostly distinct from generic video skills with only minor overlap risk, matching the anchor for 4 rather than the clearly-distinct 5.	4 / 5
	Total	15 / 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: anbeime/skill
Path: skills/video-creation-pro/video-creation-pro/SKILL.md
Commit: d9e888d

Reviewed: 2 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.