Security audit
Security checks for vulnerabilities and agentic risk
This skill appears to be a simple Chinese social-content template generator, with some overstated documentation but no evidence of hidden access, exfiltration, persistence, or destructive behavior.
Install only if you want a Chinese-language, template-based social media copy generator. Treat unsupported features in the README, such as image input, rewrite, batch mode, trend tracking, API access, and paid tiers, as documentation claims rather than working reviewed functionality.
该描述与代码有部分重合：代码确实用于自动生成多平台内容，尤其支持小红书、抖音、公众号文案生成，并提供若干发布相关建议（如配图、BGM、拍摄建议）。但存在明显的能力描述不实或不完整。首先，代码只接收 platform 和 topic 两个主要输入，没有任何图片处理流程，因此“图生文案”不成立。其次，代码是基于内置模板随机拼装内容，没有接收已有文案并进行“爆款改写”的机制。再次，声明中的输出格式较统一，而实际是按平台输出不同字段，部分平台缺少标题、标签或正文。最后，代码还支持微博和朋友圈，属于声明未覆盖的额外平台能力。综合看，主方向相近，但描述高估并部分失真了实际功能，应判定为不匹配。
Without declared permissions the skill's intent is opaque and cannot be validated.
The manifest states the skill supports '图生文案' and '爆款改写' and implies automatic output across multiple creation modes. However, the only implemented script functions accept a text topic and emit fixed template-based content; there is no image input handling, rewrite logic, or multi-mode processing in the code example at L148-L287.
The natural-language description presents the skill entirely as a Chinese content-creation assistant for Chinese platforms, with no indication that users may choose another language or locale. This can violate language/locale policy when a skill implicitly enforces one language without offering a user choice or documenting a justified region-specific restriction.
The trigger phrases are broad and overlap with common user requests such as '生成内容', 'content', and 'create post', which can cause the skill to activate unexpectedly in unrelated conversations. Over-broad activation can redirect benign user tasks into this skill's behavior, increasing the risk of unintended prompt capture, wrong-task execution, or policy bypass through skill hijacking in multi-skill environments.
The description says the skill automatically outputs '标题 + 正文 + 标签 + 发布建议'. In code, Douyin output includes hook/content/CTA/BGM but no tags, and WeChat returns only a title and outline without body text, tags, or publishing suggestions; even Xiaohongshu lacks an explicit publishing suggestion field.
The script’s top-level documentation and command usage are written only in Chinese, and the rest of the program emits Chinese-only generated content for all supported platforms. This imposes a specific language/locale on users without any opt-in or documented region-specific justification.
The usage documentation shows commands for '--input素材.md', 'rewrite', and 'batch' operations, suggesting these behaviors are available. The documented script below only parses positional arguments for platform and topic and has no file input, rewrite, or batch-processing logic, so the documentation overstates what the code does.
Skill's behavior or capabilities extend beyond its stated purpose. Scope creep allows an agent to perform actions unrelated to its documented functionality, increasing the attack surface.
copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
No suspicious patterns detected.