ClawHub
Back to skill

Security audit

eatsomething-by

Security checks for vulnerabilities and agentic risk

Overview

This skill stores diet preferences and pantry notes locally for meal recommendations, with no executable code or network behavior found.

Before installing, be aware that the skill may create local files containing dietary restrictions, preferences, goals, and food inventory. Avoid storing details you would not want kept in the skill directory, and use clear instructions when adding or removing inventory items.

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
  • 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
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
Findings (2)

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The skill explicitly instructs storage of sensitive personal dietary profile data, including allergies, intolerances, habits, and goals, into a persistent local file without any user-facing notice about retention, scope, or privacy implications. Even if the data is not highly regulated by default, it is health-adjacent personal data and can expose sensitive attributes or create safety risks if retained or accessed unexpectedly.

Missing User Warnings

Low
Confidence
87% confidence
Finding
The skill directs automatic deletion of inventory entries when food is marked as consumed, but does not require explicit confirmation or warn the user that persistent data will be removed. This creates a risk of unintended data loss or inaccurate inventory state, especially if user intent is ambiguous or the agent misinterprets a conversational statement as an update command.