ClawHub
Back to skill

Security audit

GOG Stale Game Cleanup

Security checks across malware telemetry and agentic risk

Overview

The skill does what it says, but it sends local game details by email, creates reminders by default, and its shell script has unsafe environment-variable handling that should be reviewed before use.

Install only if you are comfortable running a local shell script that reads your GOG library file, emails a report, and writes Apple Reminders. Run with DRY_RUN=true first, verify EMAIL_TO and EMAIL_ACCOUNT, avoid untrusted values in environment variables, and consider patching the script to pass data into Python via arguments or environment reads instead of interpolating into Python source.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
95% confidence
Finding
The skill invokes shell-based capabilities (`bash scripts/stale_games.sh`) and interacts with external tools, but no explicit permissions are declared in the skill metadata. That creates a transparency and policy-enforcement gap: a user or platform may not realize the skill can execute shell commands, send email, and modify reminders until after invocation.

Vague Triggers

Medium
Confidence
86% confidence
Finding
The trigger phrases include broad terms like 'game cleanup' and 'unused games,' which could cause the skill to activate in routine conversation without clear user intent. Because the skill sends email and creates reminders, accidental invocation can produce unwanted external actions and data disclosure.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The description explains functionality but does not prominently warn that the workflow sends email and writes to Apple Reminders, both of which affect external services and user data. Without an explicit warning and consent boundary, users may not understand that invoking the skill causes outbound communication and persistent reminders to be created.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The script automatically emails a report containing installed game names and local install paths, which can reveal sensitive usage and filesystem information to the configured recipient without any explicit confirmation, minimization, or redaction step. Even if intended as a convenience feature, install paths may expose usernames, directory structure, external volumes, or other personal context, making unintended disclosure more harmful than a simple activity summary.

VirusTotal

58/58 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.