ClawHub

FGO Invoicing

Security checks across malware telemetry and agentic risk

Overview

The skill is a coherent FGO invoicing helper, but it needs Review because it can mutate real invoice records and debug mode can expose sensitive business data.

Install only if you are comfortable giving the agent access to FGO production credentials and invoice mutation commands. Use UAT for testing, avoid FGO_DEBUG or --debug with real customer data, and require explicit human approval before issuing, cancelling, deleting, or reversing invoices.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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)

Lp3

Medium
Category
MCP Least Privilege
Confidence
86% confidence
Finding
The skill clearly uses sensitive capabilities (environment secrets, file input, and outbound network access) but does not declare explicit permissions. That weakens platform-level review and user understanding, and can allow a skill with privileged behavior to run with less scrutiny than intended. In this context the risk is elevated because the skill handles an API private key and performs high-impact financial actions against an external invoicing service.

Missing User Warnings

Medium
Confidence
98% confidence
Finding
When debug mode is enabled, the CLI logs full request URLs, headers, and form bodies to stderr. Those bodies include sensitive fields such as CodUnic, computed Hash values, and full invoice/customer/payment data; response logging may also expose returned invoice metadata and links. In automation environments, stderr is often centrally collected, making credential and business-data leakage realistic.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal