Security audit

Cheat Code Paid

Security checks for vulnerabilities and agentic risk

Overview

This skill is not clearly malicious, but it mixes broad coding assistance with enterprise knowledge retrieval while requesting command, file, API, cache, and shared-storage authority.

Review before installing. Use it only with clearly approved knowledge sources, avoid submitting secrets or sensitive code unless your organization permits those destinations, and keep cache/shared directories out of repositories or sync folders unless the contents are intended for that audience.

SkillSpector

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
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
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access

Findings (5)

Intent-Code Divergence

High

Confidence: 92% confidence
Finding: The skill is documented as an enterprise knowledge-retrieval tool, but its parameters and output schema instead describe code-review scoring and security/compliance grading. This capability mismatch can cause the agent to invoke the skill in the wrong context, leading users to send source code or sensitive material to a tool they believed was only performing retrieval, increasing the risk of unintended processing or disclosure.

Intent-Code Divergence

Medium

Confidence: 86% confidence
Finding: The document states cached data remains local, but later instructs users to place shared knowledge directories under version control or sync tooling. This creates a misleading trust boundary: users may store sensitive query results believing they never leave the machine, when in practice they can be propagated to repositories or shared drives.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The activation language is broad enough to match ordinary development and coding-assistance requests, not just narrowly scoped knowledge retrieval. Over-broad triggering increases the chance the skill activates on unrelated prompts and gains access to read/exec/write capabilities or external data flows in situations where the user did not intend to use it.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill describes writing to local cache and shared storage paths, including team-shared directories, without a prominent upfront warning that it will create and persist files. Users may unknowingly store sensitive query results, internal knowledge, or derived data on disk or in synchronized folders, creating confidentiality and retention risks.

Missing User Warnings

Medium

Confidence: 91% confidence
Finding: The skill promotes querying external knowledge services and custom data sources but does not clearly warn users that internal prompts, documents, or code may be transmitted to external APIs or internal systems. In an enterprise context, this omission is dangerous because users may submit confidential material under the assumption of purely local processing.

Static analysis

No suspicious patterns detected.