Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only product validation planning skill with broad activation wording but no unsafe execution, persistence, or data-access behavior.
Installers should understand that this skill may activate on broad product or startup wording, so it is best suited for product validation planning rather than general coding, operations, or business advice. It does not appear to run commands, access private data, or persist changes.
The invocation guidance lacks clear boundaries on when the skill should and should not be used, and the listed keywords span very broad business and startup topics. This makes accidental or excessive invocation more likely, which can override more suitable skills or produce irrelevant advice in ordinary conversations about business, validation, or prototypes.
The trigger sentence is extremely broad and maps to common user phrasing like asking for help or a practical workflow, which can cause the skill to activate in contexts far beyond its intended product-validation scope. Over-broad activation increases the chance of unintended routing, misleading outputs, or the skill intercepting unrelated requests without clear user intent.
The trigger phrases are broad and generic enough to match many ordinary user requests about product ideas, workflows, validation, or startups. This can cause unintended invocation of the skill, increasing prompt-routing risk and making it easier for the skill to capture requests outside its narrow intended scope.
The skill description is broad enough to activate on many generic business, product, or implementation-related requests, increasing the chance of unintended routing. Overly broad activation can cause the wrong skill to take over conversations, leading to irrelevant guidance, policy bypass through skill misselection, or reduced reliability of downstream agent behavior.
The trigger keywords include highly ambiguous terms like 'prototype', 'validation', 'saas', and 'startup', which are common across many unrelated requests. This makes accidental invocation more likely and can misroute users into a skill that assumes a product-validation workflow even when their request concerns coding, operations, or general brainstorming.
The skill is written entirely in Chinese and does not provide any user language negotiation or fallback behavior, which can cause the agent to respond in an unexpected language. While not a direct code-execution issue, forced language context can confuse users, degrade reviewability, and make it harder to detect unsafe or incorrect outputs if the user's expected language differs.
The trigger keywords are broad, generic business terms like 'validation', 'prototype', 'saas', and 'startup', which can easily match ordinary user conversation and cause unintended skill activation. Over-broad activation can misroute user requests, override more appropriate skills, and increase the attack surface for prompt injection or unsafe workflow execution when the skill is invoked outside its intended scope.
The default prompt uses a very broad invocation phrase tied to a wide range of user intents, which can cause the skill to be invoked in contexts that were not clearly intended by the user. Because implicit invocation is also enabled, this increases the chance of over-triggering, misrouting user requests, or exposing the skill's instructions in unrelated conversations.
The activation description uses very general business and product terms without clear boundaries, making the skill eligible for many everyday requests that are only loosely related. This weak specificity increases the chance of unintended invocation and expands the attack surface for prompt-routing abuse or inappropriate delegation.
The trigger sentence is broad enough to match normal user phrasing about getting help or needing a workflow, which can cause the skill to activate outside its intended scope. Over-broad activation can route unrelated conversations into this skill, leading to incorrect handling, prompt hijacking opportunities through unintended invocation, and reduced trust in skill boundaries.
This file is a Chinese-language README, but the usage section presents trigger phrases only in English and does not indicate whether users may interact in Chinese or choose their preferred language. That can amount to an implicit language constraint without opt-in.
