ClawHub
Back to skill

Security audit

Hong Kong Supermarket Shopping

Security checks for vulnerabilities and agentic risk

Overview

This skill appears benign: it looks up supermarket prices by downloading and caching a public CSV, with the cache behavior disclosed and limited to its own data folder.

Before installing, users should be comfortable with the skill contacting the Consumer Council pricewatch site and storing a short-lived CSV cache inside the skill directory. The packaging inconsistency in openclaw.plugin.json may need correction for compatibility, but I did not find evidence of hidden access, exfiltration, destructive behavior, or unsafe persistence.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • 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)

Natural-Language Policy Violations

Low
Confidence
88% confidence
Finding
The manifest states that the skill auto-detects the query language and responds in that language. This imposes a language/locale behavior without offering the user a choice or opt-in, which matches the policy category for language or locale constraints.

Missing User Warnings

Low
Confidence
87% confidence
Finding
This code fetches a CSV from the network, deletes prior cached CSV files, and writes a new file to disk. Although it logs these actions to stderr, there is no user-facing disclosure in the normal output or prompt that the skill performs network access and local file modification, which falls under missing user warnings for code files.

Static analysis

No suspicious patterns detected.