Security audit

Work Productivity Nano Banana Workflow Helper

Security checks for vulnerabilities and agentic risk

Overview

This workflow helper provides ordinary planning guidance, with broad activation keywords but no sensitive access or automatic high-impact actions.

Expect general planning and checklist guidance. Prefer explicit invocation for Nano Banana workflow tasks because generic activation keywords may select it for unrelated requests.

A.I.G

Vulnerability Patterns

Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

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 (6)

Activation keywords include broad everyday terms without Nano Banana-specific constraints.

Medium

Category: Trigger Abuse
Confidence: 98% confidence
Finding: The activation keywords include generic terms such as “pro,” “generate,” “edit,” and “bug fix” without requiring Nano Banana context, potentially causing this skill to activate for unrelated requests. The metadata further broadens its scope to general workflows and implementation support, increasing the risk of inappropriate routing and irrelevant guidance if the host uses these terms for skill selection. This README contains no malicious execution instructions or exfiltration behavior, so the demonstrated risk is limited to activation scope and reliability.
Content: No source excerpt is available for this finding.

Activation keywords include generic terms that can trigger the skill outside its intended Nano Banana workflow context.

Medium

Category: Trigger Abuse
Confidence: 97% confidence
Finding: The usage section lists standalone keywords such as pro, generate, edit, images, and bug fix, which overlap with many unrelated requests. It provides no exclusion conditions or requirement that these keywords occur within a Nano Banana-specific request; the more specific example trigger sentences do not constrain the keyword list.
Content: No source excerpt is available for this finding.

Activation keywords include common words without a Nano Banana-specific scope requirement.

Medium

Category: Trigger Abuse
Confidence: 99% confidence
Finding: Common activation terms such as "pro", "generate", and "edit" are not restricted to Nano Banana-related requests, potentially causing this skill to activate on unrelated tasks and displace more appropriate workflows. This is a low-impact routing weakness rather than evidence of malicious behavior: the supplied content contains ordinary planning guidance and no instructions for exfiltration, destructive actions, or security bypasses.
Content: No source excerpt is available for this finding.

Generic standalone keywords can activate the skill outside Nano Banana workflows.

Medium

Category: Trigger Abuse
Confidence: 99% confidence
Finding: L44 lists broad activation keywords such as pro, generate, edit, and images, which occur frequently in unrelated requests. L04 similarly allows generic keywords without requiring Nano Banana context, and the examples provide no exclusions to prevent unintended invocation.
Content: No source excerpt is available for this finding.

Activation keywords include broad everyday terms without a Nano Banana Pro context requirement.

Medium

Category: Trigger Abuse
Confidence: 94% confidence
Finding: The activation keywords include generic terms such as "pro," "generate," "edit," and "images" without requiring Nano Banana Pro context, potentially causing this skill to activate for unrelated requests and displace more appropriate workflows. The metadata reinforces this overly broad scope, but the supplied file contains only generic planning guidance and no apparent harmful execution or exfiltration instructions, limiting the demonstrated impact to routing and output integrity.
Content: No source excerpt is available for this finding.

Implicit invocation is enabled, but the description does not define a clear activation scope.

Low

Category: Trigger Abuse
Confidence: 88% confidence
Finding: Implicit invocation is enabled while the skill metadata uses broad triggers such as 'nano', 'banana', and 'pro', and the interface description does not establish a precise task boundary. This creates a low-severity activation-scope weakness: unrelated requests could invoke the skill and introduce irrelevant workflow guidance. The supplied file contains no executable actions, sensitive access requirements, or malicious instructions, so there is no evidence of a higher-impact exploit.
Content: No source excerpt is available for this finding.