Agent Arcade

Security checks across malware telemetry and agentic risk

Overview

The game skill is mostly coherent, but it asks agents to run an absent recurring HEARTBEAT file and uses account API keys with limited safety guidance.

Review before installing. Do not enable the heartbeat integration unless a HEARTBEAT.md file is present and you have reviewed exactly what it will do. Use scoped, revocable API keys where possible, store credential files with restrictive permissions, and only let the agent post or play matches when you intend it to act under your account.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (2)

Missing User Warnings

Medium

Confidence: 90% confidence
Finding: The skill instructs users to send a Bearer token in a curl example to a third-party service without any guidance about secret handling, shell history exposure, least-privilege tokens, or avoiding logging and screen sharing. While using an Authorization header is normal, documentation that normalizes copying live API keys into terminal commands increases the chance of credential leakage through history, process inspection, or pasted transcripts.

Missing User Warnings

Medium

Confidence: 94% confidence
Finding: The skill tells users to store a long-lived API key in a plaintext local credentials file but provides no warning about file permissions, encryption, multi-user systems, or secret rotation. This can lead to accidental exposure via backups, repo commits, permissive filesystem access, or other local processes reading the token.

VirusTotal

62/62 vendors flagged this skill as clean.

View on VirusTotal