ClawHub
Back to skill

Security audit

Wishlist

Security checks for vulnerabilities and agentic risk

Overview

This wishlist skill is a simple, disclosed personal tracking workflow with expected file storage and price-checking behavior.

Installers should expect the skill to create and update Markdown files in ~/wishlist and, when price checking is used, to query external stores or search sources using wishlist item details. Review stored wishlist contents if they include sensitive personal preferences.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • 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)

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
- Review monthly: still want it?

## What NOT To Do
- Buy impulsively without checking wishlist first
- Keep items forever without reviewing
- Ignore price history patterns
- Forget why you wanted something
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Missing User Warnings

Low
Confidence
93% confidence
Finding
The skill instructs the agent to search current prices across external stores, which implies outbound network requests and possible transmission of user interest data, but it does not tell the user that external queries will occur or request consent. In this wishlist context the data is not highly sensitive by default, yet purchase interests and links can still reveal preferences or habits, so the privacy risk is real though limited.

Static analysis

No suspicious patterns detected.