Security audit

Cheat Code

Security checks for vulnerabilities and agentic risk

Overview

This skill has no executable payload, but it requests broad read/write/command authority while using vague “limitless” and “hidden ability unlock” framing that is not safely scoped.

Review carefully before installing. Treat this as a broadly privileged development skill: only use it in a controlled workspace, avoid sensitive files or production systems, and require explicit confirmation before any file write, API credential use, or command execution.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (8)

Intent-Code Divergence

Medium

Confidence: 86% confidence
Finding: The skill documents a generic content-processing input/output contract while elsewhere advertising automation, file operations, API use, and command execution. This mismatch obscures the real behavior of the skill, making it easier for a user or host agent to invoke higher-risk actions than the interface description suggests.

Intent-Code Divergence

Medium

Confidence: 91% confidence
Finding: The document claims commands run in a safe sandbox and that only whitelisted commands are allowed, but provides no concrete enforcement details. Unsupported safety claims are dangerous because they can cause users and orchestrators to trust a skill with exec capability that may actually run arbitrary or insufficiently constrained commands.

Vague Triggers

High

Confidence: 89% confidence
Finding: The description uses broad trigger language such as making the agent 'limitless' and telling it whatever you want, which can match routine user requests and encourage overbroad activation. In a skill with read/write/exec tools, vague invocation language increases the chance of the agent applying powerful capabilities outside narrowly intended contexts.

Vague Triggers

High

Confidence: 88% confidence
Finding: The main capabilities section presents activation in vague, everyday language rather than precise operational boundaries. This makes accidental or opportunistic invocation more likely and is especially risky because the skill advertises privileged actions like code generation and execution-related workflows.

Missing User Warnings

Medium

Confidence: 84% confidence
Finding: The markdown describes file handling, API integration, and command execution but does not provide prominent user-facing warnings about data exposure, system modification, network egress, or operational risk. In a skill that can read, write, and execute, missing warnings reduce informed consent and can lead to unsafe use with sensitive files or production systems.

Ssd 1

Medium

Confidence: 93% confidence
Finding: The marketing language explicitly frames the skill as unlocking hidden or limitless agent abilities, which is a red flag because it encourages behavior beyond normal guardrails. In security review, language implying bypass of constraints is dangerous even without code because it socially engineers users and agents to treat policy boundaries as optional.

Ssd 1

High

Confidence: 95% confidence
Finding: The application scenarios explicitly promote 'hidden capability unlocking' as a use case, suggesting unauthorized expansion of agent behavior rather than a legitimate, bounded function. Given that the skill also exposes read/write/exec-adjacent functionality, this context makes the finding more dangerous because it normalizes misuse of powerful actions.

Ssd 1

Medium

Confidence: 90% confidence
Finding: The feature list repeats claims of enabling hidden or limitless agent behavior instead of describing bounded functionality. Repeated framing of constraint bypass increases the likelihood that users interpret the skill as a mechanism to evade safeguards, which is especially risky in combination with execution and file-system capabilities.

Static analysis

No suspicious patterns detected.