ClawHub

Paddle

Security checks across malware telemetry and agentic risk

Overview

This Paddle skill is a coherent payments-integration guide, with expected billing and local-memory behavior that users should still handle carefully.

Install this only if you are using Paddle for billing work. Use sandbox keys first, keep API keys and webhook secrets in environment variables or a secret manager, require explicit review before any live subscription changes, and keep ~/paddle/memory.md limited to non-secret integration notes.

SkillSpector

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

Vague Triggers

Medium
Confidence
95% confidence
Finding
The instruction "Should I help whenever you're working on payments, or only when you ask?" creates an overly broad activation rule that can cause the skill to engage on loosely related future tasks without clear per-request consent. In an agent setting, ambiguous auto-activation increases the chance of unintended handling of sensitive billing, subscription, or payment-adjacent workflows and can override user expectations about when the skill is active.

Missing User Warnings

Medium
Confidence
97% confidence
Finding
The skill directs the agent to persist conversation-derived preferences in ~/paddle/memory.md without instructing it to notify the user, obtain consent, or minimize what is stored. Persistent storage of environment details, product/pricing information, and business preferences can expose sensitive operational context across sessions and create privacy and data-governance issues if the file is later accessed by other tools or skills.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal