Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable code, persistence, credential handling, or hidden data access, though its trigger wording is overly broad.
This skill appears safe to install from a security perspective, but users should be aware that its broad implicit triggers may activate it for unrelated productivity or image-generation requests. A maintainer should narrow the trigger terms to explicit Nano Banana Pro workflow requests to improve routing reliability.
The trigger sentence is broad and phrased like normal user language, which increases the chance the skill is invoked when the user did not explicitly intend to call it. In an agent ecosystem, unintended invocation can misroute tasks, override more appropriate skills, or cause confusing workflow execution based on weak keyword matches.
The trigger phrases are excessively generic and include common natural-language requests such as 'Help me' and 'I need a practical workflow,' which can cause the skill to activate outside its intended scope. In an agent ecosystem, this creates prompt-routing confusion and increases the chance the skill will intercept unrelated user requests, leading to inappropriate execution paths or accidental delegation.
The manifest description says to use the skill when a user asks for broad terms like "work-productivity," "nano," "banana," or "pro," which creates an overly permissive invocation condition. In agent routing systems, such broad matching can incorrectly select this skill for benign unrelated tasks, increasing attack surface and degrading safety controls based on intended skill boundaries.
The trigger keyword list includes extremely generic terms such as "nano", "banana", "pro", "generate", "edit", and "images", which can match many unrelated user requests. This can cause the skill to activate outside its intended scope, leading to misrouting, unintended prompt injection exposure, or suppression of a more appropriate skill.
The top-level description says the skill should be used when users ask for broad terms like work-productivity, nano, banana, or pro, without clearly defining exclusion conditions or activation thresholds. This ambiguity can make the skill match a wide range of unrelated requests, increasing accidental invocation and causing the agent to follow the wrong workflow or produce irrelevant outputs.
The trigger keywords include extremely broad everyday terms such as "nano", "banana", "pro", "generate", and "edit", which can cause the skill to activate for unrelated user requests. Over-broad invocation increases the chance that this skill intercepts conversations outside its intended scope, leading to incorrect routing, policy mismatches, or unintended behavior in downstream agent workflows.
The trigger sentence is extremely broad and can match ordinary user language unrelated to this specific skill, causing unintended activation. Overbroad routing increases the chance that the skill intercepts requests outside its intended scope, which can lead to wrong-task execution, user confusion, or misuse of downstream automation in contexts the user did not request.
The invocation guidance is ambiguous because it mixes broad keywords such as 'nano', 'banana', 'pro', 'generate', and 'bug fix' without clear boundaries or exclusions. This can cause accidental activation for unrelated requests, making skill routing unreliable and potentially interfering with safer or better-matched skills.
Natural-language policy checks apply to all file types. A file labeled as Simplified Chinese contains substantial English operational text, including the requirement description, workflow description, and trigger phrases, which can impose a language expectation without explicit opt-in or explanation.
The example trigger phrases are broad and effectively teach the router or user to invoke the skill with minimal contextual grounding. While not directly executable, these examples reinforce over-triggering behavior and can cause the skill to be selected for vague or unrelated prompts.
该文件是 zh-CN 技能说明，但 L04、L11、L13 保留了大段英文内容与英文用户群描述，未说明应根据用户语言偏好切换输出，容易让技能在中文语境中默认夹带英文说明。若技能据此强制或优先使用特定语言而未征得用户同意，可能构成语言/locale 政策问题。
The skill’s display text and default prompt are extremely broad and include generic terms like 'work-productivity', 'practical workflow', 'checklist', 'analysis', and 'implementation support', which can cause unintended invocation for ordinary user requests unrelated to the intended niche workflow. Because implicit invocation is enabled, this ambiguity increases the chance the agent activates in contexts where it was not specifically requested, potentially causing prompt-routing errors, user confusion, or unintended access to the skill’s specialized instructions.
The activation scope is ambiguous because the listed keywords and trigger sentences are loosely defined and overlap with many normal productivity or image-generation requests. In a skill-routing system, this can cause accidental invocation and prompt hijacking of unrelated tasks, reducing reliability and potentially exposing users to outputs, assumptions, or automations that were not appropriate for their request.
