Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with broad auto-invocation wording, but no evidence of hidden code, credential access, persistence, exfiltration, or destructive behavior.
Installers should be aware that this skill may activate on broad productivity or bug-fix language. Prefer invoking it by its exact name when you want this workflow, and consider narrowing its triggers if your environment supports implicit skill routing.
The description says to use the skill when a user asks for broad categories like work-productivity, self, improving, or proactive, which makes invocation criteria overly permissive. In a multi-skill agent environment, this increases accidental routing and can cause the agent to select this skill for unrelated tasks, reducing reliability and potentially bypassing safer or more specialized workflows.
The trigger list includes extremely generic terms such as "self," "improving," and "proactive," which are likely to appear in many unrelated user requests. This can cause unintended skill invocation, leading the agent to apply the wrong workflow, override a more appropriate skill, or expose users to unrequested autonomous behavior patterns.
The trigger list includes extremely broad everyday terms such as "self", "improving", and "proactive", which are likely to match many unrelated user requests. This can cause accidental activation of the skill in inappropriate contexts, leading to prompt hijacking of routing behavior, degraded agent reliability, and unintended disclosure or modification of workflow outputs.
The trigger phrases are extremely broad and include common terms like 'self', 'improving', and 'proactive', which can cause the skill to activate for many unrelated user requests. In an agent ecosystem, overbroad routing can misapply this skill's workflow to unintended contexts, increasing the chance of incorrect automation, user confusion, or bypass of more appropriate specialized skills.
The trigger sentence is broad enough to match ordinary help requests rather than a narrowly scoped security-reviewed use case. That can cause unintended auto-invocation of the skill, leading the agent to apply self-improving or proactive workflow behavior in contexts the user did not explicitly request, increasing the chance of prompt-scope confusion or unsafe side effects.
The invocation description uses ambiguous natural language that does not clearly define when the skill should or should not activate. In an agent ecosystem, this ambiguity can make the skill compete with unrelated skills or fire on common productivity requests, which is risky because the skill is framed as proactive and self-improving and may influence downstream agent behavior unexpectedly.
The file is presented as a Chinese-language README and includes Chinese labels and usage guidance, but there is no natural-language note offering users a language choice or explaining that the locale is intentionally region-specific. Under the stated policy, forcing a specific language without opt-in can be a policy violation unless clearly justified.
The trigger phrases are extremely broad, including generic terms like "self", "improving", and "proactive", which can cause the skill to activate for many ordinary user requests unrelated to the intended workflow. In an agent environment, this creates prompt/skill-routing collisions that may override more appropriate skills, causing unintended behavior, lower reliability, and increasing the attack surface for indirect prompt steering.
The example trigger sentences reinforce activation on generic phrasing like "Help me" and "I need a practical workflow," which trains routing logic or implementers toward vague matching behavior. While not directly executable, this increases the chance of over-triggering and misclassification in downstream systems that rely on examples to shape invocation behavior.
该技能文件为 SKILL.zh-CN.md，正文也全部使用简体中文，但文件中没有说明这是用户可选的语言版本，也没有提供语言选择或切换说明。若技能因此默认以特定语言响应，可能构成语言/locale 强制要求而缺少用户 opt-in。
The skill description defines applicability in very broad terms, including generic categories like work-productivity and practical support artifacts, without clear exclusion criteria. Ambiguous activation boundaries increase the chance that the skill will be selected for unrelated tasks, which can distort agent behavior and reduce safety by applying the wrong instructions to the wrong user intent.
The manifest enables implicit invocation without defining narrow triggers or exclusions, so the skill may be auto-selected in contexts only loosely related to its purpose. Because this skill is broadly framed around productivity and self-improving workflows, the loose scope increases the chance of inappropriate invocation, unintended prompt influence, or unsafe execution paths in unrelated conversations.
The file references Chinese-language evidence sources and titles, but it does not state whether the skill operates in Chinese, English, or offers a language choice. If this content implies a language expectation for use or activation, the lack of explicit user opt-in can create a locale-policy concern.
