AC aikey-review
AI KEY·发布前审一遍 ——口播稿发布前审核技能。按观众的四次决定审：点不点进来 · 留不留下来 · 记不记得你 · 做不做点什么。先盲读开头判留走，逐句信息密度评分（60/80 分线）+ 十一问 + 六种反应预测（只算观众走得到的部分，含流量是谁、离生意多远）+ 红线五查（改法给稳妥版和保留力度版两版）+ 机器信号层（导流 / 广告形状 / 名单词，与内容违规分开报），默认只诊断不改。不负责重写整稿（→ aikey-script）、换开头或标题（→ aikey-hook / aikey-title）、拆别人的爆款（→ aikey-resonate）。 触发方式：/aikey-review、/能不能发、/审核、/aikey-审核、「这稿子能不能发」「帮我审一下」「过一遍红线」「信息密度够不够」 Pre-publish review for talking-head scripts, organised around the viewer's four decisions: click, stay, remember, act. Per-sentence density scoring, eleven questions, red-line audit. Diagnose-only by default. Not for: rewriting the script (aikey-script), replacing the opening or title (aikey-hook / aikey-title), decoding someone else's hit (aikey-resonate). Trigger: /aikey-review, "can I publish this", "review my script" —— AI KEY · 不给公式，给判据。每条规则都标了实测代价。
AI KEY·发布前审一遍 ——口播稿发布前审核技能。按观众的四次决定审：点不点进来 · 留不留下来 · 记不记得你 · 做不做点什么。先盲读开头判留走，逐句信息密度评分（60/80 分线）+ 十一问 + 六种反应预测（只算观众走得到的部分，含流量是谁、离生意多远）+ 红线五查（改法给稳妥版和保留力度版两版）+…
As a process C 57/100 · Has gaps — weak spots: result and completion, inputs and preconditions, failures and branches
How to improve
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 5. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- note
frontmatter-keyunknown frontmatter key "slug" - note
frontmatter-keyunknown frontmatter key "displayName"
Process rating: all ten parameters 57/100
- 0Result and completion. Does not say what the result is
- 0Inputs and preconditions. Does not say what the process needs to start
- 0Failures and branches. Linear process with no failure handling
- 0Progress reporting. Says nothing while it works
- 50When it triggers. No condition that starts the skill
- 100Tools and files. No external tools needed
- 100Steps. 53 steps
- 100Consistency. Name and required fields are in place
- 100Execution cost. Instruction body is 2575 tokens
- 100Running it twice. No mutating operations
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +3Description length 812: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- -231 emoji in the instructions: noise for the model
- +1No license
- +2Single-language instructions
- +5Description quotes 2 example trigger phrases
- +4Description says when NOT to use the skill
- +4Structure: 15 headings
- +3Step-by-step instructions: 53 items
- +4Has examples (1 code blocks)
- +4Reference files are cited in the instructions (3 of 3)
Quality base 70; lint remarks subtract, signals add up to 100. Result: 90.