ClawHub
Back to skill

Security audit

hk-insurance-news

Security checks for vulnerabilities and agentic risk

Overview

This skill is a straightforward Chinese-language generator for a Hong Kong insurance weekly WeChat-style market report, with expected web research and image/link inclusion.

Before installing, expect the skill to search public Hong Kong insurance sources and produce Chinese WeChat-style Markdown with links and possible images. Review all embedded links, image URLs, and any local file references before publishing or sharing the generated article.

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
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • 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)

Natural-Language Policy Violations

Medium
Confidence
94% confidence
Finding
The skill instructions and output expectations are entirely specified in Chinese, including generation of a WeChat article, without indicating that language choice is user-selectable. This can violate language/locale policy when a skill implicitly forces a specific language without opt-in.

Missing User Warnings

Low
Confidence
94% confidence
Finding
The skill instructs the agent to include local paths or remote images/links in the generated Markdown without any user warning or trust boundary guidance. This can cause the agent to surface untrusted remote content or reference sensitive local files, creating risks of privacy leakage, tracking, or accidental disclosure when the document is rendered or shared.

Static analysis

No suspicious patterns detected.