CtrlK
BlogDocsLog inGet started
Tessl Logo

chengfeng-export-talking-head

把已审核的口播删词账本落成剪后视频:展示确认卡、执行物理剪切、验收剪后媒体。只在删词已经人工复核之后使用。用户说导出口播、执行剪切、把剪好的视频导出来、继续导出,或确认卡回传 action=continue_cut 时使用。不要用于生成删词候选、字幕、分镜动画或成片渲染。

74

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

85%

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

An exceptionally well-structured, actionable, and defensive workflow with strong validation checkpoints for an irreversible media cut. The two soft spots are some extra justificatory prose and external reference paths whose target files are absent from the bundle.

Suggestions

Tighten the narrative passages (e.g. the '整条链子上第一个产生媒体文件' framing and the revision-substitution rationale) into one-line rules so every token is an executable instruction.

Provide the referenced bundle files (business-workflow-contract.md, runtime-and-product-contract.md) under ./references/ — or inline the one-level-deep detail — so the progressive-disclosure navigation resolves to real files.

Consider moving the '恢复与失败' enumeration into a short reference table while keeping the three most safety-critical cases inline, to reduce body length without losing the failure-loop discipline.

DimensionReasoningScore

Conciseness

The prose is well-ordered and avoids teaching generic concepts, but several explanatory passages (e.g., the framing of '这是整条链子上第一个产生媒体文件的地方' and the repeated philosophical justification of why substituting revisions is dangerous) add token weight beyond the strictly executable contract.

2 / 3

Actionability

Each step ships concrete, executable bash invoking real scripts (ensure-runtime, ensure-running, workflow get, cuts apply) with exact flags and a revision-frozen cuts-apply command, making the whole skill copy-paste runnable.

3 / 3

Workflow Clarity

A numbered 0–4 sequence with explicit precondition assertions, a confirmation-card checkpoint, a post-confirmation revision re-check, and dedicated '恢复与失败' recovery loops — exactly the validate→fix→retry discipline the rubric rewards for destructive operations.

3 / 3

Progressive Disclosure

The body references two external files ([业务 Skill 的阶段合同] and [Runtime 与产品契约]) but no references/, scripts/, or assets/ bundle exists in the skill directory, so those paths are not verified-loadable; structure is clear but disclosure cannot be confirmed against a real bundle.

2 / 3

Total

10

/

12

Passed

Description

100%

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, trigger-rich, and answers both what and when while explicitly scoping the skill out of adjacent tasks. Its only mild weakness is that several triggers assume the domain workflow context, but nothing is vague or padded.

DimensionReasoningScore

Specificity

Lists several concrete actions — '展示确认卡、执行物理剪切、验收剪后媒体' — naming a domain (talking-head 删词 export) with distinct executable steps, matching the 'multiple specific concrete actions' anchor.

3 / 3

Completeness

Clearly answers both what ('落成剪后视频' via the three steps) and when ('只在删词已经人工复核之后使用' plus explicit triggers and an exclusion clause), satisfying the explicit-both-what-and-when anchor.

3 / 3

Trigger Term Quality

Natural user phrasings are explicit and varied — '导出口播、执行剪切、把剪好的视频导出来、继续导出' — plus the structured `action=continue_cut` trigger, giving good coverage of terms users would actually say.

3 / 3

Distinctiveness Conflict Risk

A narrow niche with distinct triggers and an explicit negative-scope clause ('不要用于生成删词候选、字幕、分镜动画或成片渲染') that fences it off from neighboring skills, making wrong-skill triggering unlikely.

3 / 3

Total

12

/

12

Passed

Validation

93%

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

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 suspicious

Warning

Total

15

/

16

Passed

Repository
Agentchengfeng/chengfeng-videocut-skills
Reviewed

Table of Contents

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.