Security audit

Voidex Areana Space Trading

Security checks across malware telemetry and agentic risk

Overview

This appears to be a real game skill, but it asks the agent to search broadly for credentials and store its API key persistently without tight limits.

Install only if you are comfortable with an autonomous game agent using a Voidex Arena account. Provide VOIDEX_ARENA_KEY explicitly, do not allow the agent to search arbitrary files, notes, or .env files for credentials, avoid storing the API key in general memory or unencrypted config files, and use a pseudonymous owner handle if possible.

SkillSpector

By NVIDIA

Vulnerability Patterns

Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (6)

Context-Inappropriate Capability

Medium

Confidence: 96% confidence
Finding: The skill explicitly instructs the agent to search memory, notes, local files, config files, .env files, and environment variables for credentials. For a game skill, this is overbroad secret discovery behavior that can cause unrelated sensitive tokens to be accessed or reused outside the immediate task context.

Context-Inappropriate Capability

Medium

Confidence: 95% confidence
Finding: The skill tells the agent to persistently store the API key in environment variables, config files, or memory systems for future reuse. Persisting secrets beyond the session expands the attack surface, increases accidental disclosure risk, and is not justified for a recreational trading-game integration.

Missing User Warnings

Medium

Confidence: 93% confidence
Finding: The credential search instructions lack any sensitivity boundaries or warnings, encouraging indiscriminate inspection of memory, files, and environment data. In an agent setting, that can normalize broad secret harvesting and lead to privacy violations or exposure of unrelated credentials.

Missing User Warnings

Medium

Confidence: 88% confidence
Finding: The registration flow requires users to submit an `ownerHandle` tied to a personal Twitter/X account, but the documentation provides no privacy notice, purpose limitation, retention guidance, or safer alternative. This creates unnecessary collection of personal data and can expose operators or users to privacy, profiling, or doxxing risk, especially in a competitive public leaderboard setting.

Ssd 3

High

Confidence: 98% confidence
Finding: Broadly directing the agent to collect and reuse credentials from memory, notes, files, and environment risks sweeping in sensitive secrets unrelated to the game. Because the skill context is low-stakes entertainment, this behavior is especially unjustified and materially increases the chance of credential misuse or cross-context data exposure.

Ssd 3

Medium

Confidence: 96% confidence
Finding: Instructing long-term retention of the API key and agent name in memory or files encourages unnecessary secret persistence. Persistent storage makes later compromise, unintended reuse, or leakage through other tools more likely, especially when the skill does not define secure storage requirements.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.