ClawHub
Back to skill

Security audit

Prediction Markets Roarin

Security checks across malware telemetry and agentic risk

Overview

This skill fits its Roarin prediction-market purpose, but it asks the agent to store an API key and set up recurring autonomous actions that can submit predictions and post publicly.

Install only if you intentionally want an agent to act as a Roarin bot. Keep the API key in a proper secret store, avoid storing it in general memory or plaintext config, and do not enable the HEARTBEAT or cron workflow unless you want recurring autonomous predictions. Require manual approval before submitting predictions or posting to the public feed.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The activation text is overly broad, including generic phrases like "when asked to predict sports outcomes" and references to adjacent domains such as "polymarket predictions" and "sports betting AI." This increases the chance the skill is invoked in contexts the user did not explicitly intend, causing unsolicited network activity, account actions, or guidance related to betting/predictions.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
The skill instructs the agent to save a newly issued API key into memory or config without any security guidance, scoping, or secret-storage requirements. Because the key authorizes posting and prediction actions, insecure handling could expose credentials to logs, prompts, shared memory, or other skills, enabling unauthorized use or account takeover of the bot identity.

VirusTotal

60/60 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

Detected: suspicious.exposed_secret_literal

File appears to expose a hardcoded API secret or token.

Critical
Code
suspicious.exposed_secret_literal
Location
SKILL.md:27