ClawHub
Back to skill

Security audit

World Cup and Football Predictor

Security checks across malware telemetry and agentic risk

Overview

This skill is a football forecasting tool with disclosed local scripts and local record-keeping, with no evidence of exfiltration, credential access, or hidden destructive behavior.

Before installing, understand that using the helper scripts may create local forecasting and review ledgers under ~/.football-forecaster unless you pass --no-record or choose another --data-dir. Do not provide proprietary match notes or sensitive internal data unless you are comfortable storing them locally, and treat all betting forecasts as analysis rather than guarantees.

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 (3)

Lp3

Medium
Category
MCP Least Privilege
Confidence
91% confidence
Finding
The skill declares no permissions, yet its instructions explicitly direct use of local scripts, file paths, and shell commands such as `python3 scripts/*.py` and reading/writing data under `~/.football-forecaster`. That creates a capability/permission mismatch: an agent or reviewer may treat the skill as low-risk while the content operationally expects code execution, filesystem access, and possibly environment interaction.

Description-Behavior Mismatch

Medium
Confidence
82% confidence
Finding
The CLI writes forecast records to disk by default unless the caller explicitly disables it, which can surprise users who expected pure analysis behavior. In environments where match inputs may include proprietary data, identifiers, or operator notes, this creates an unnecessary local data-retention surface and can leak information to shared disks, backups, or later readers.

Missing User Warnings

Low
Confidence
76% confidence
Finding
Writing records to disk without an explicit disclosure at the point of write weakens user awareness and consent, especially because the default path is chosen automatically when no output path is supplied. This is primarily a transparency and privacy-control issue rather than a direct exploit primitive, but it can still cause unintended retention of potentially sensitive operational data.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.