ClawHub
Back to skill

Security audit

健康科普文案台

Security checks for vulnerabilities and agentic risk

Overview

This is a text-only health education copywriting skill with disclosed medical-safety limits and no code, persistence, credential access, or data exfiltration behavior.

Reasonable to install for drafting health education content. Users should still review medical claims, avoid using it for individualized diagnosis or treatment advice, and be aware that broad trigger phrases may activate the skill during general health-copy editing conversations.

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
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • 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)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger examples are broad generic phrases such as health-popularization and medical-copy requests, which can match many ordinary user prompts and cause the skill to activate when the user did not explicitly intend to invoke it. In practice this can override normal assistant behavior and steer unrelated health discussions into the skill’s template, increasing the chance of inappropriate or premature medical-content generation in a sensitive domain.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
90% confidence
Finding

The natural-language examples are phrased as ordinary conversational follow-ups, including generic editing requests like shortening or turning content into a table. These can be mistaken for skill invocations during routine conversation, leading to accidental continued use of the skill workflow even when the user is just asking for normal assistant help.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.