Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only skill for improving error messages, with no evidence of hidden execution, data access, persistence, or destructive behavior.
Installers should be aware that the trigger wording is broad and may activate the skill for general debugging or support requests. The artifact is otherwise low risk: review or tighten triggers if precise routing matters in your environment.
The trigger sentence begins with a very broad phrase ('Help me'), which can cause the skill to activate for many ordinary requests that are only loosely related to error-message improvement. Overbroad activation increases the chance of unintended invocation, prompt routing mistakes, and the skill being applied outside its intended scope where it may produce irrelevant or unsafe guidance.
The activation guidance is too unspecific and does not define strong boundaries for when the skill should or should not run. That ambiguity can lead to accidental invocation during unrelated debugging, support, or productivity conversations, increasing misrouting risk and reducing the reliability of agent behavior.
The trigger sentence is phrased as a broad natural-language request that overlaps with ordinary user conversation, which can cause the skill to activate unintentionally. In an agent ecosystem, overly generic activation increases the chance of prompt-routing mistakes, causing the wrong skill to handle user data or produce actions outside the user's intended workflow.
The trigger phrase is so generic that it can match ordinary user requests about help, workflows, support, or error messages, causing the skill to activate in situations the user did not explicitly intend. Over-broad activation increases the chance of inappropriate routing, prompt/context injection into unrelated tasks, and accidental disclosure or processing of sensitive troubleshooting content.
The skill description includes very broad activation terms like work-productivity, debugging, support, and analysis, which can match many ordinary user requests unrelated to this specific skill. This can cause unintended invocation or over-selection of the skill, leading to prompt-routing errors, inappropriate context injection, or interference with more suitable skills.
The trigger keywords are generic conversational terms that appear frequently in normal interactions, so they do not safely distinguish intentional use of this skill from routine discussion. In a skill-routing system, this increases the chance of accidental activation and can degrade security boundaries by applying the wrong workflow or exposing unnecessary capability context.
The example trigger sentences use natural everyday phrasing that could easily appear in ordinary requests, without clear boundaries showing when this skill should or should not activate. That makes the examples functionally overbroad and reinforces ambiguous routing behavior, which can produce incorrect tool selection or unnecessary skill exposure.
L04 表述为当用户提出 work-productivity, error messages, debugging, user feedback, support，或需要围绕该需求获得流程、产物、检查清单、分析或实现支持时使用，其中前半部分是宽泛主题，后半部分几乎覆盖任意工作产出请求。这种描述没有清楚区分“错误信息改进”专属场景与一般生产力/支持类请求，触发边界模糊。
L44 将 debugging、support、user feedback 等宽泛主题词直接作为触发关键词，但没有限定上下文、边界或排除条件。这些词在大量普通对话中都可能出现，容易让技能在并非针对“错误信息改进”场景时被调用。
The default prompt uses a very broad, natural-language invocation phrase that overlaps with common user requests about error messages and debugging. Because implicit invocation is enabled, this increases the chance the skill is triggered unintentionally during ordinary conversation, causing prompt injection of the skill’s instructions or unexpected routing into the skill without clear user intent.
The trigger scope is ambiguous because it maps a wide range of general productivity and support language to this skill without enough constraints on user intent or input type. In practice, this can lead to misrouting unrelated conversations into the skill, reducing reliability and potentially causing the agent to request or transform sensitive troubleshooting context when the user did not intend to use this workflow.
该技能文件为 SKILL.zh-CN.md，内容整体以简体中文呈现，但未看到向用户说明语言选择、切换机制或这是面向特定语言受众的显式限定。按规则，若技能强制特定语言而无用户选择或合理、清晰的适用说明，可能构成语言/locale 策略问题。
No suspicious patterns detected.
