Security audit

Work Productivity Nano Banana Workflow Helper

Security checks across malware telemetry and agentic risk

Overview

This is a documentation-only workflow helper with overly broad activation wording, but no evidence of hidden execution, credential use, persistence, or data exfiltration.

This skill appears safe to install for Nano Banana Pro-style workflow help, but its trigger wording should be tightened before broad distribution. Users or maintainers should prefer explicit invocation or more specific Nano Banana workflow phrases to reduce accidental activation in unrelated editing, image, or bug-fix requests.

SkillSpector

By NVIDIA

Vulnerability Patterns

Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (10)

Vague Triggers

Medium

Confidence: 88% confidence
Finding: The trigger sentences use very generic terms such as 'work-productivity', 'nano', 'banana', 'pro', 'generate', 'edit', and 'bug fix', which can cause the skill to activate for unrelated requests. Over-broad activation increases the chance of unintended invocation, confusing routing, and accidental exposure of users to instructions or workflows they did not request.

Vague Triggers

High

Confidence: 92% confidence
Finding: The trigger keywords and example activation phrases are extremely broad, including generic terms like "pro," "generate," "edit," and "bug fix." This can cause the skill to activate outside its intended scope, leading to unintended interception of unrelated user requests and potentially overriding more appropriate skills or workflows.

Vague Triggers

High

Confidence: 97% confidence
Finding: The trigger list includes extremely generic terms like "nano," "banana," "pro," "generate," and "edit," which are likely to appear in many unrelated user requests. This can cause accidental skill activation, expanding the skill's effective authority and making routing less predictable, especially in environments where multiple skills compete for invocation.

Vague Triggers

High

Confidence: 94% confidence
Finding: The manifest description says to use the skill for broad concepts like "practical workflow, artifact, checklist, analysis, or implementation support," which lacks meaningful boundaries and overlaps with many ordinary tasks. In a skill-selection system, this ambiguity can cause overmatching and unintended routing, potentially diverting unrelated requests into this skill's instructions.

Vague Triggers

High

Confidence: 96% confidence
Finding: 触发关键词包含“nano”“banana”“pro”“generate”“edit”“images”“gemini”“bug fix”等高频或多义词，边界非常宽，容易在与该技能无关的普通对话中被意外激活。误触发会让代理偏离用户真实意图，注入不相关流程或文件内容，降低可预测性，并可能放大其他有害指令的执行机会。

Vague Triggers

Medium

Confidence: 90% confidence
Finding: 描述中的激活条件写成“当用户提出 work-productivity, nano-banana-pro, nano, banana, pro，或需要…支持时使用”，范围过宽且没有清晰排除条件，等于把大量泛化需求都纳入触发面。这样的定义会造成技能路由歧义，使系统在普通生产力、图片编辑或修复问题场景下错误选择该技能。

Vague Triggers

Medium

Confidence: 87% confidence
Finding: 示例触发句使用了非常常见的求助模板，如“Help me”“I need a practical workflow”，但没有提供区分性边界，容易与海量正常用户请求重叠。示例本应起到约束作用，这里反而扩大了技能的匹配面，增加误调用和行为混淆风险。

Vague Triggers

High

Confidence: 93% confidence
Finding: The default prompt contains highly generic trigger terms such as "work-productivity," "practical workflow," "checklist," and "analysis," which overlap with common user requests. Combined with skill routing, this can cause the skill to activate in situations the user did not specifically intend, increasing the chance of prompt-scope confusion and unintended instruction injection into unrelated tasks.

Vague Triggers

High

Confidence: 96% confidence
Finding: Enabling implicit invocation without strict trigger constraints allows the platform to auto-select this skill based on vague semantic matches rather than explicit user intent. In this skill, the broad productivity-oriented language makes accidental activation more likely, which can expose users to unintended behavior, context contamination, or execution of irrelevant workflow guidance in sensitive conversations.

Vague Triggers

High

Confidence: 90% confidence
Finding: The trigger keywords and example phrases are overly broad and include generic terms like "work-productivity," "nano," "banana," and "pro," which can match many unrelated user requests. This can cause accidental invocation of the skill in contexts it was not intended for, leading to irrelevant guidance, prompt-surface expansion, and potential misuse if the skill is selected during sensitive workflows.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.