Security audit
Security checks for vulnerabilities and agentic risk
This is a documentation-only workflow helper with overbroad auto-invocation triggers but no code execution, credential access, persistence, or destructive behavior.
Before installing, consider narrowing or disabling implicit invocation because the current trigger list could make this skill appear during unrelated image-editing, Gemini, bug-fix, or generic productivity requests. The package otherwise appears low-risk: it is documentation-only and does not ask for credentials, network execution, persistence, or system-level access.
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 ambiguous invocation boundary. In agent ecosystems, overly permissive descriptions can cause incorrect auto-selection or over-triggering, reducing reliability and potentially steering unrelated user tasks into this skill's workflow.
The trigger list is excessively broad, including generic terms like 'nano', 'banana', 'pro', 'generate', 'edit', and 'bug fix' that can appear in many unrelated requests. This can cause the skill to activate outside its intended scope, leading to prompt hijacking of routing, confused task selection, and accidental exposure of users to instructions or workflows that do not match their request.
The trigger keyword list includes very broad generic terms such as "nano", "banana", "pro", "generate", "edit", and "images", which are likely to appear in unrelated user requests. This can cause accidental activation of the skill outside its intended scope, leading to incorrect routing, prompt-context pollution, and increased risk that the agent applies workflow instructions in inappropriate contexts.
The trigger sentence is extremely broad and reads like ordinary user language, which increases the chance the skill will be invoked unintentionally during unrelated conversations. In an agent environment, overbroad auto-invocation can route user data or workflow control into the wrong skill, causing confusion, unsafe task execution, or prompt-scope interference.
The trigger phrases are extremely broad, relying on generic terms like 'nano', 'banana', 'pro', 'generate', and 'edit', which can cause the skill to activate for many unrelated user requests. In an agent ecosystem, this increases the chance of unintended invocation, context hijacking, or the skill influencing workflows it was not meant to handle, reducing predictability and potentially exposing users to incorrect or unsafe automation paths.
This file is a Chinese-language README, but it does not indicate that language selection is optional or provide any user opt-in for locale preference. Under the stated policy, forcing a specific language without choice can be a natural-language policy violation unless the locale constraint is documented and justified.
The trigger phrases are very broad, including generic terms like "nano," "banana," "pro," "generate," and "edit," which can cause the skill to activate in contexts unrelated to its intended workflow. This creates skill-routing ambiguity and increases the chance the agent applies the wrong workflow, producing misleading guidance or interfering with safer, more appropriate skills.
The skill description says it should be used when users mention broad categories like work-productivity, nano, banana, or pro, or when they need a practical workflow, checklist, analysis, or implementation support. Those conditions are too open-ended and make the activation boundary unclear, which increases the chance of the skill being invoked for many unrelated tasks.
The example trigger phrases are truncated and highly generic, so they do not establish meaningful specificity for when the skill should activate. Poor examples reinforce ambiguous matching behavior and make it harder for maintainers or routing logic to distinguish intended from unintended invocations.
The default prompt is overly broad and includes generic terms like workflow, artifact, checklist, analysis, and implementation support, which can cause the skill to be selected for requests outside its intended scope. This increases the chance of unintended invocation, prompt-routing confusion, and unsafe delegation of unrelated user tasks into this skill.
Enabling implicit invocation without strong activation constraints allows the platform to auto-select this skill based on loose semantic matches rather than clear user intent. In combination with the broad productivity-oriented description, this can lead to accidental invocation, context leakage into the skill, and unexpected execution paths in multi-skill environments.
The invocation description lacks a clear boundary and matches a broadly phrased request pattern, making accidental activation likely for many benign productivity requests. Because this skill is framed as a general helper for practical workflows, ambiguous triggering could cause it to preempt more appropriate skills and produce irrelevant or risky actions under incorrect assumptions.
