ClawHub
Back to skill

Security audit

OOTD

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward outfit recommender that uses weather plus disclosed local style and wardrobe files, with no evidence of hidden execution or unsafe behavior.

Install only if you are comfortable with the skill using your local style profile and optional wardrobe file to personalize recommendations. Avoid putting sensitive information in USER.md or wardrobe.json beyond what you want used for outfit advice.

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
  • Rogue AgentSelf-Modification, Session Persistence
  • 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)

Session Persistence

Medium
Category
Rogue Agent
Content
### 1. The Digital Wardrobe (Optional)

Create a file named `wardrobe.json` in your workspace (`~/.openclaw/workspace/`) to define your clothes. If this file doesn't exist, the skill will give general advice based on your style.

**Example Format:**
```json
Confidence
60% confidence
Finding
Skill establishes unauthorized persistence across sessions via cron jobs, startup scripts, or state files. Session persistence allows an attacker to maintain access beyond the current interaction.

Missing User Warnings

Low
Confidence
89% confidence
Finding
The skill explicitly states it reads `USER.md` and may read `wardrobe.json`, both of which can contain sensitive personal information such as preferences, habits, and owned items, but it does not present an explicit privacy notice or consent boundary. In a personal assistant context this is not inherently malicious, but it creates a privacy risk because users may not realize the extent of local data access involved in generating recommendations.

Static analysis

No suspicious patterns detected.