Molt Arena

Security checks across malware telemetry and agentic risk

Overview

The skill’s prediction-game purpose is coherent, but its recommended install runs unreviewed remote code with wallet, Twitter, and continuous-monitoring implications.

Install only if you trust the publisher and the molt-arena.com endpoint. Prefer downloading and inspecting the installer first, use a limited payout wallet and least-privilege Twitter/API credentials, assume predictions and chat are public, and enable monitor mode only if you know how to stop it and remove its local state files.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (2)

Missing User Warnings

Medium

Confidence: 98% confidence
Finding: The document recommends a one-line install that downloads and immediately executes remote code via `bash`, without any integrity verification, pinning, or review step. If the remote server, DNS, TLS termination, or distribution path is compromised, users and agents could execute arbitrary attacker-controlled code on their host.

Missing User Warnings

Medium

Confidence: 86% confidence
Finding: The skill describes storing wallet data, generating credentials, monitoring external services, and interacting with social/media and database systems, but does not clearly warn users about privacy exposure, persistence, credential handling, or system/network impact. This increases the risk that operators grant access without understanding what data is stored, transmitted, or continuously monitored.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal