Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with no executable payload, though its broad trigger wording could make it activate unexpectedly.
Before installing, be aware that this skill may trigger on ordinary requests involving words like generate, edit, images, pro, or bug fix. It appears safe from an execution and privilege perspective, but maintainers should narrow the trigger language to avoid unintended activation.
The trigger phrases are broad enough to match many ordinary requests, including generic terms like 'nano', 'pro', 'generate', 'edit', and 'bug fix'. This can cause the skill to activate unintentionally, leading to incorrect routing, unexpected behavior, or the skill influencing conversations outside its intended scope.
The manifest description says the skill should be used when a user asks for broad terms like 'work-productivity', 'nano', 'banana', or 'pro', or generally 'needs a practical workflow, artifact, checklist, analysis, or implementation support'. This creates an ambiguous activation boundary that could make the skill eligible for a wide range of unrelated requests, increasing accidental invocation and inappropriate instruction injection into other contexts.
The trigger list includes extremely generic terms such as 'nano', 'banana', 'pro', 'generate', 'edit', 'images', and 'bug fix', which are likely to match many unrelated user requests. Over-broad activation can cause the wrong skill to engage, leading to prompt hijacking of task routing, irrelevant instructions being injected into benign workflows, and unintended expansion of this skill's authority.
The trigger list includes extremely broad, high-frequency terms such as "nano", "banana", "pro", "generate", and "edit", which can cause this skill to activate for many unrelated user requests. Overbroad activation can route users into the wrong workflow, increasing the chance of unsafe or irrelevant guidance and making the agent easier to steer unintentionally.
The trigger sentence is overly broad and can cause the skill to activate for generic requests containing common words like 'help me' or partial requirement text rather than a clear request for this specific workflow. That increases the chance of incorrect routing, unintended invocation, and context confusion, which can degrade safety if users are steered into an irrelevant skill during sensitive tasks.
The invocation guidance does not define firm scope boundaries, so the skill may be selected for loosely related requests involving work productivity, bug fixes, implementation help, or generic analysis. Because this skill is positioned as a helper for a broad job-to-be-done, weak constraints make accidental activation more likely and can degrade safety, predictability, and user intent matching across the agent environment.
The trigger examples are broad enough to match ordinary productivity requests containing generic words like 'help me' or requests for a 'practical workflow,' which can cause the skill to activate outside its intended niche. In an agent ecosystem, this increases the chance of unintended routing, overshadowing more appropriate skills and causing users to receive actions or guidance from the wrong workflow.
The Chinese README contains substantial English instructional content, which can impose a language expectation inconsistent with the localized document. Because the file does not indicate that users may choose their preferred language, this creates a locale/language policy concern.
The example triggers are vague, repetitive, and truncated, and they do not demonstrate clear invocation boundaries or realistic user intent. Poor trigger examples reinforce overbroad matching behavior and make it easier for unrelated prompts to be treated as in-scope, though the examples themselves are less dangerous than the manifest and keyword list.
The description says the skill should be used when users mention broad product-like terms or need generic artifacts such as workflows, checklists, analysis, or implementation support, but it does not define clear boundaries for when the skill should not activate. This ambiguity can cause misrouting and unintended invocation, especially because the named terms overlap with common language and adjacent tasks.
The default prompt contains a very broad natural-language trigger phrase covering common terms like 'help me' and generic workflow/support language, which can cause unintended invocation during ordinary user conversation. Because implicit invocation is enabled, this raises the chance that the skill activates when the user did not specifically intend to call it, potentially injecting its behavior or context into unrelated tasks.
Although the sentence includes some context, it still remains broad enough to match many unrelated requests because it starts with a generic phrase and embeds a long, reusable requirement description. This can cause accidental skill selection and prompt-routing errors, especially in systems that match loosely on trigger text or keywords.
