ClawHub
Back to skill

Security audit

World Cup Assist Value Trading

Security checks across malware telemetry and agentic risk

Overview

This skill is transparent about most of what it does, but it contains dormant code that can place real trades with limited safeguards, so it needs review before use.

Install only if you intend to use Simmer and Tavily credentials for a market-signal tool and understand that changing DRY_RUN can make it place real, potentially irreversible trades. Keep it in dry-run or paper mode unless you have restricted credentials, position limits, and explicit review of every live run.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (4)

Lp3

Medium
Category
MCP Least Privilege
Confidence
93% confidence
Finding
The skill documentation indicates use of environment variables (`SIMMER_API_KEY`, `TAVILY_API_KEY`) and outbound network access to Simmer, Tavily, and third-party stats sources, but no corresponding permissions are declared. This creates a capability/permission mismatch that can mislead operators and prevent proper sandboxing or policy review, increasing the risk of secret exposure or unintended external requests when the linked script is run.

Description-Behavior Mismatch

High
Confidence
98% confidence
Finding
The skill advertises itself as signal-printing only, but it contains live trade execution code guarded only by a mutable boolean flag. If DRY_RUN is changed, intentionally or accidentally, the script can place real trades, creating financial risk and violating user expectations about the skill's behavior.

Intent-Code Divergence

High
Confidence
99% confidence
Finding
The runtime output states that the tool only prints signals and does not place trades, yet the code includes a path that can execute trades. This mismatch is dangerous because operators may trust the message and run the tool in contexts where any trading capability is unacceptable, leading to accidental financial actions.

Missing User Warnings

Medium
Confidence
89% confidence
Finding
The notes explicitly outline a path from research to live deployment, including replacing dry-run behavior with actual Simmer API calls and scheduling regular runs, but provide no warning that this would automate financial trading with real-world consequences. In an agent skill focused on market signals, that omission increases the chance that a user or downstream agent operationalizes the strategy without understanding the financial risk, execution risk, or need for safeguards.

VirusTotal

61/61 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.