메뉴
Performs three-level risk scanning for platform compliance (absolute claims, false advertising, sensitive content). Use when reviewing marketing copy, social media content, or any text for regulatory compliance. Trigger phrases: "违禁词检测", "合规审查", "风险扫描", "广告法"
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
从文字脚本自动生成专业视频故事板（Word-to-VideoStory）— 基于结构化提示词工程 + 多模态AI工具链，将任意文字脚本转化为带角色参考、分镜画面、故事板文档和专业排版的全套输出。触发词："故事板""分镜""storyboard""视频脚本可视化"
Vibe Engineering Discipline v1.0 — 8 principles for disciplined AI-assisted "Vibe Coding". Use this skill when starting a new vibe-coding project, setting ground rules with an AI coding agent, or onboarding a team to structured AI collaboration.
MINDRING 从0到1增长体系 — 冷启动SOP、内容输出SOP、百元验证法、商业九步模型、品牌出海策略、B2B经营模型、流量四层模型。
AI赋能独立创业者的5阶段作战体系 — 市场洞察、战略落地、执行管理、极速开发、冷启动与全球化验证。基于3个月跑通全球化线上服务的实战经验提炼。
Four communication strategies based on knowledge overlap (I know you don't, you know I don't, we both don't, we both know). Use when designing AI interaction, teaching communication, or optimizing human-AI dialogue. Trigger phrases: "4种交流方式", "认知差异", "沟通策略", "知识重叠"
Evaluates business project feasibility using 5 basic questions. Use when assessing startup viability, business model validation, or investment due diligence. Trigger phrases: "商业项目预判", "创业可行性", "商业模式", "5个基本问题"
| name | prohibited-words-detection |
| description | Performs three-level risk scanning for platform compliance (absolute claims, false advertising, sensitive content). Use when reviewing marketing copy, social media content, or any text for regulatory compliance. Trigger phrases: "违禁词检测", "合规审查", "风险扫描", "广告法" |
Act as a senior platform compliance reviewer. Scan given text with three-level risk detection and output results in the specified format.
Input: "这款面膜全网销量第一！医院皮肤科都在用，7天淡化所有皱纹，添加的xxx成分能修复受损细胞，私信我送独家护肤秘籍"
Output format:
🚨【违规词1】"全网销量第一"
- 类型：绝对化用语（违反《广告法》第9条）
- 建议改为："累计热销超10万盒"
🚨【违规词2】"医院皮肤科都在用"
- 类型：医疗关联暗示
- 建议改为："实验室实测数据表明"
📊 综合风险等级：🔴 红色（需修改N处）
User input: "帮我检测这段文案是否合规" AI action: Scans text, identifies violations, provides replacement suggestions Expected result: Structured compliance report with risk level