Security audit

game-library

Security checks across malware telemetry and agentic risk

Overview

This is a coherent game-library helper, but a crafted search term could make it run unintended code.

Review before installing. The skill appears intended to query your own Steam and Epic libraries, but it relies on logged-in local tools and sends some game identifiers to external services for name enrichment. Avoid using it with untrusted or automatically generated search terms until the search implementation is fixed to pass input safely.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands

Findings (5)

Lp3

Medium

Category: MCP Least Privilege
Confidence: 93% confidence
Finding: The skill declares and documents capabilities that include shell execution, network access, and reading/writing local cache files, but it does not declare permissions or present clear guardrails around those operations. This weakens user and platform visibility into what the skill can do, increasing the risk of unintended command execution, external data access, or local file modification without informed consent.

Description-Behavior Mismatch

Medium

Confidence: 94% confidence
Finding: The steam command forwards arbitrary user-supplied arguments directly to the external Steam CLI, bypassing the skill's intended normalized query interface. This can expose broader operations than list/search/stats and may allow unintended actions or disclosure depending on what the installed steam CLI supports.

Vague Triggers

Medium

Confidence: 81% confidence
Finding: The trigger description is broad enough to activate on many ordinary game-related requests, which can cause the agent to invoke this skill when the user did not intend to access linked Steam/Epic accounts or run local CLI commands. In context, this is risky because the skill can query account-linked libraries and perform networked lookups, so over-triggering can expose private game-library data or cause unnecessary external requests.

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill states it will scrape a third-party site to enrich Chinese names, but it does not clearly warn users that game identifiers or titles from their library may be sent to an external service outside Steam/Epic. In this context, even if the transmitted data is 'only' game metadata, it can still reveal ownership patterns or preferences and creates privacy, compliance, and reliability risks from unsanctioned scraping.

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: The user-controlled search term is interpolated directly into inline Python source inside double-quoted shell strings. A crafted term containing quotes and Python syntax can break out of the intended string literal and execute arbitrary Python code in the context of the script.

VirusTotal

VirusTotal findings are pending for this skill version.

View on VirusTotal

Static analysis

No suspicious patterns detected.