Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed product-validation planning workflow with no code execution, credential access, persistence, or destructive behavior.
This skill appears safe to install for product validation planning. Be aware that it may activate on broad startup, SaaS, validation, or prototype requests, so use explicit skill selection if you want tighter control.
The trigger phrases are extremely broad and map to common business/product language, which can cause the skill to activate for ordinary user requests that were not meant to invoke it. Over-broad routing increases the chance of prompt/skill hijacking at the orchestration layer, unintended data exposure to the skill context, and incorrect task handling because the wrong skill is selected.
The trigger phrases are broad and loosely scoped, making it easier for the skill to activate on generic product, validation, workflow, or startup-related requests that may not actually need this specific skill. Overbroad activation can cause inappropriate routing, unintended prompt injection surface expansion, and reduced user control over which skill is invoked, especially in an agent ecosystem where skills may have powerful downstream actions or instructions.
The trigger phrases are broad enough to match common requests about workflows, validation, prototypes, or startup help, which can cause the skill to activate outside narrowly intended scenarios. In an agent ecosystem, overbroad activation can route unrelated user requests into this skill, increasing the chance of inappropriate guidance, prompt-scope confusion, or unintended use of downstream capabilities.
The skill description is broad enough to match many ordinary business, product, and implementation requests, which can cause the agent to invoke this skill in situations where it was not specifically intended. Overly generic routing increases the chance of unintended prompt capture, response shaping, and interference with more appropriate skills or base-model handling.
The trigger keywords and example phrases are extremely generic terms such as 'validation,' 'prototype,' 'saas,' and 'startup,' which commonly appear in benign conversation. This makes accidental or excessive invocation likely, allowing the skill to inappropriately steer responses across a wide range of unrelated user requests.
The description says to use the skill for broad categories like business-and-operations, product idea, validation, prototype, and SaaS without defining boundaries or exclusion cases. This ambiguity increases the chance of accidental invocation in unrelated business discussions, which can distort agent behavior and reduce reliability of skill selection.
The trigger keywords are broad business terms such as "validation," "prototype," and "startup," which can overlap with many ordinary conversations and cause the skill to activate outside its intended niche. Over-broad activation can route users into an unsuitable workflow, leading to confused outputs, unintended prompt shaping, or bypass of more appropriate domain-specific skills.
The skill enables implicit invocation and pairs it with a very broad default prompt and broad usage description, which can cause the agent to activate in situations only loosely related to the user’s intent. This increases the chance of over-triggering, unintended routing, and prompt-surface expansion where the skill influences conversations outside its intended scope.
This file is labeled as a Chinese README, but several core instructional sections, including the requirement summary, workflow description, and trigger phrases, are written in English. That creates an implicit language/locale inconsistency without telling users whether English is required or optional.
该文件为 SKILL.zh-CN.md，整体内容固定为中文（简体，中国区域）版本，但文中未说明这是用户可选语言，也未提供语言切换或适用区域约束。按自然语言政策要求，若技能强制特定语言/locale 而无用户选择，可能构成语言/区域策略问题。
The file includes multiple Chinese-language evidence titles alongside English content, but provides no indication of expected output language, user language choice, or locale handling. While not necessarily harmful by itself, this can create an implicit language-policy ambiguity if the skill surfaces or relies on language-specific material without user opt-in.
