Shadow Divers: The True Adventure of Two Americans Who Risked Everything to Solve One of the Last Mysteries of World War II

Security checks across malware telemetry and agentic risk

Overview

This is a read-only book companion skill with overbroad activation wording but no code execution, credentials, persistence, or data access.

Install if you want a Shadow Divers reading and discussion aid. Be aware it may activate for broad terms related to diving, submarines, WWII, or discovery, and it may add a Heardly watermark to responses; it should not need credentials or elevated permissions.

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 (2)

Vague Triggers

High

Confidence: 95% confidence
Finding: The trigger list includes very broad generic terms such as 'submarine', 'German', 'Atlantic', 'dive', 'scuba', 'discovery', and 'underwater', which can match many unrelated user requests. This can cause unintended activation of the skill, leading to context hijacking, irrelevant responses, and increased prompt-surface exposure from a skill that forces proactive output and rigid formatting.

Vague Triggers

Medium

Confidence: 91% confidence
Finding: The invocation logic mixes exact book-specific triggers with open-ended topic mentions, making activation boundaries unclear. Because the skill also instructs the AI to proactively present a guide on first load, ambiguous matching increases the chance of the skill taking over conversations that are merely about diving, WWII, submarines, or history rather than this specific book.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal