ClawHub
Back to skill

Security audit

Harrypotter

Security checks across malware telemetry and agentic risk

Overview

This is a narrow Harry Potter lookup skill that discloses public HP-API use and shows no sensitive access, persistence, or destructive behavior.

Reasonable to install for public Harry Potter lookups. Be aware that using it may send lookup terms to HP-API, and if you install from the external GitHub repository referenced in the README, inspect the executable scripts there because they were not included in this reviewed artifact.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Missing User Warnings

Low
Confidence
91% confidence
Finding
The README explicitly states that the skill uses a third-party API, but it does not warn users that running the CLI will make outbound network requests to an external service. In an agent-skill context, undisclosed network access can surprise operators, leak usage metadata such as queries and IP address, and violate least-expectation or network-restriction policies even when the API itself is harmless.

Vague Triggers

Medium
Confidence
82% confidence
Finding
The activation guidance is broad enough that an agent may invoke this skill for general Harry Potter questions without strong gating or user intent confirmation. While this skill appears low-risk because it only performs unauthenticated lookups to a public API, over-broad invocation can still cause unnecessary external requests, incorrect tool selection, and unintended disclosure of user queries to a third-party service.

VirusTotal

66/66 vendors flagged this skill as clean.

View on VirusTotal

Static analysis

No suspicious patterns detected.