ClawHub
Back to skill

Security audit

Privacy Cards

Security checks across malware telemetry and agentic risk

Overview

This documentation-only skill is purpose-aligned, but it enables sensitive financial card operations without enough guardrails for irreversible changes or card-data exposure.

Review before installing. Use sandbox where possible, keep the API key least-privileged, require explicit confirmation before creating, updating, or closing cards, verify the exact card token before any destructive action, prefer pausing over closing when reversible suspension is sufficient, and avoid exposing full card numbers, CVVs, expiry values, transaction details, or API keys in chats, logs, terminal history, or shared files.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (4)

Missing User Warnings

Medium
Confidence
86% confidence
Finding
The skill includes a permanent destructive action (`state: CLOSED`) and labels it as permanent, but it does not instruct the caller to obtain explicit user confirmation or warn about operational consequences such as irreversible loss of card usability. In a payments-management skill, omission of confirmation guidance increases the risk of accidental destructive changes through automation or user misunderstanding.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The example response displays highly sensitive cardholder data fields such as `pan`, `cvv`, and expiration values without a strong privacy or handling warning. Even if the values are illustrative, normalizing inclusion of full payment credentials in skill output can encourage unsafe logging, copying, display to unintended recipients, or downstream prompt leakage in agent workflows.

Missing User Warnings

Medium
Confidence
91% confidence
Finding
The reference explicitly documents fields for full PAN, CVV, and expiry values, which are highly sensitive payment card data, but provides no warning about PCI-sensitive handling, least-privilege access, logging restrictions, or redaction requirements. In a skill designed to create and manage virtual cards, this omission increases the chance that downstream agents or users will request, display, store, or log full card credentials unsafely.

Missing User Warnings

Medium
Confidence
84% confidence
Finding
The update-card documentation notes that CLOSED is permanent, but does not present this as an explicit irreversible-action warning or require caution around destructive state changes. In an agent skill that can automate card management, weak warning language raises the risk of accidental permanent closure, causing service disruption and forcing card reissuance or manual recovery steps.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.