ClawHub
Back to skill

Security audit

pangolinfo-amazon-daily-competitor-radar

Security checks across malware telemetry and agentic risk

Overview

This is a coherent Amazon competitor monitoring skill, with disclosed local snapshots and optional daily scheduling that users should opt into deliberately.

Install this if you are comfortable giving the skill access to Pangolinfo-backed Amazon market data tools and storing per-ASIN snapshots locally. Do not enable the daily schedule unless you want recurring runs, credit usage, and repeated product lookups; review the cron or CronCreate entry after enabling it.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Context-Inappropriate Capability

Medium
Confidence
91% confidence
Finding
The skill instructs the agent to create recurring scheduled jobs (`CronCreate`) or provide automation templates, which expands behavior from one-time competitor analysis into persistence and repeated autonomous execution. That creates a control-boundary issue: an agent could establish ongoing tasks, repeated data access, or noisy automation without explicit, per-run user consent and without this capability being declared in the skill’s core scope.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill directs the agent to read and write baseline snapshots under `~/.pangolinfo/baselines/<asin>.json` without an explicit user warning or consent flow for local filesystem persistence. Even though the stored fields are mostly product metadata, silent local writes introduce statefulness, privacy/retention concerns, and an unexpected side effect for a monitoring/reporting skill.

VirusTotal

57/57 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.