Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overly broad activation language, but no executable code, credential handling, persistence, or hidden high-impact behavior.
Installers should be aware this skill may be selected for some generic weather, API, or bug-fix prompts because its activation wording is broad. Review or narrow the trigger terms if precise routing matters, but the artifact itself does not show malicious or high-impact behavior.
The trigger phrase is extremely broad and resembles ordinary user speech, which can cause the skill to activate in unintended contexts. In an agent ecosystem, this creates prompt-routing confusion and may cause the wrong skill to handle requests, leading to unreliable behavior, accidental data exposure to an unnecessary workflow, or bypass of more appropriate safeguards.
The manifest description contains very broad activation terms like "work-productivity," "api," and "practical workflow" that can match many unrelated requests. In an agent-routing system, this can cause unintended invocation, incorrect tool/skill selection, and prompt-scope hijacking where this skill preempts more appropriate specialized skills.
The trigger keyword list includes highly generic words such as "api," "key," "required," "users," and "bug fix," which are common across many unrelated tasks. This makes accidental invocation likely and increases the chance that the skill captures prompts outside its intended domain, degrading routing integrity and potentially interfering with safer or more appropriate skills.
The trigger sentence begins with a broad helper phrase ('Help me ...') that overlaps with normal user language and can cause the skill to activate when a user did not specifically intend to invoke this workflow. In an agent environment, unintended invocation can route user requests into the wrong skill, creating confusion, incorrect actions, or accidental handling of unrelated tasks.
The usage section defines activation in vague terms such as broad keywords and generic helper phrasing without clear scope limits. This ambiguity increases the chance of accidental invocation and skill collision with other weather, productivity, API, or bug-fix workflows, making agent behavior less predictable and easier to manipulate through crafted prompts.
The invocation guidance lacks precise trigger boundaries and mixes generic categories with common phrases, making the activation surface ambiguous. This increases the chance of accidental invocation, especially because the skill is framed as a general workflow helper for broadly useful tasks rather than a narrowly constrained operation.
The trigger examples are very broad and can match ordinary user requests about weather, APIs, forecasts, or bug fixes without clear boundaries, causing the skill to activate in contexts the user did not intend. In an agent ecosystem, this can lead to prompt hijacking of unrelated tasks, unnecessary disclosure of context to the skill, or incorrect workflow execution.
The example trigger sentences use broad natural-language phrases and truncated requirement text without clearly stating when the skill should and should not activate. These examples can teach downstream authors or routing systems to invoke the skill for vague requests, reinforcing overbroad matching behavior.
描述中的适用范围写成“当用户提出 work-productivity, weather, current, forecasts, api，或需要围绕该需求获得实用流程…” ，边界非常宽，几乎覆盖大量普通生产力、API 或天气相关对话。过宽的激活条件会放大错误路由概率，使代理在缺乏明确上下文时调用该技能，降低系统可预测性并可能输出不适合当前任务的操作建议。
触发关键词包含大量高频、通用词，如“current”“api”“key”“users”“required”等，容易在与天气工作流无关的普通请求中误触发该技能。这会导致路由偏离用户真实意图，若技能随后生成配置、代码或安全建议，可能把不相关任务引导到错误流程，带来误操作、错误建议或信息暴露风险。
The default prompt embeds a very broad invocation phrase tied to common terms like work-productivity, weather, current, forecasts, api, workflow, checklist, and analysis. Because the skill also allows implicit invocation, ordinary user requests may unintentionally trigger this skill, causing prompt/context injection into unrelated conversations and unpredictable agent behavior.
Enabling implicit invocation without clear activation constraints increases the chance that this skill will run in contexts the user did not intend. In this skill, the broad productivity/weather wording makes that risk more pronounced, potentially leading to unauthorized tool use, irrelevant actions, or attacker-crafted phrasing that steers the agent into this workflow.
The phrase 'I need a practical workflow for ...' is generic and lacks domain-specific constraints, so many unrelated requests could match it. Because this skill is framed as a broadly useful workflow helper, the generic trigger increases the chance of accidental invocation and misrouting in multi-skill systems.
示例触发句使用了高度泛化的自然语言模板，如“Help me…”“I need a practical workflow…”，会向作者和调用方暗示只要表达模糊需求即可激活技能。这种示范会进一步弱化调用边界，增加在相似但不相关请求中被选中的概率。
