ClawHub
Back to skill

Security audit

康复训练计划台

Security checks for vulnerabilities and agentic risk

Overview

This is a coherent rehab-planning prompt skill with medical-use cautions and no code, persistence, credential access, or data exfiltration behavior.

Install only if you want a prompt helper for turning clinician-provided rehab advice into a structured plan. Because the output can affect health decisions, users should confirm any plan with a qualified clinician, provide full recovery context, and stop immediately for the red-line symptoms the skill describes.

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
  • MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
91% confidence
Finding

The trigger keywords include broad phrases such as everyday rehab-related requests that may cause the skill to activate when the user did not intend to invoke this specific medical-planning workflow. In a healthcare context, unintended activation is more serious because the system may generate structured rehabilitation guidance that users could rely on despite incomplete context or without clinician oversight.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
88% confidence
Finding

The natural-language examples are open-ended and broad enough to match routine conversation, which increases the chance of accidental invocation. Because this skill operates in a medical domain and outputs actionable exercise regimens, accidental triggering can lead to inappropriate or overconfident guidance based on partial information, making the context more dangerous than in a low-risk domain.

Content

No source excerpt is available for this finding.

Intent-Code Divergence

Low
Category
Not specified by scanner
Confidence
82% confidence
Finding

L69 says progress-stage outputs such as scheduling/record-table style enhanced capabilities belong to a Pro version, but L74's example and earlier description present items like progress recording and structured outputs as part of the standard skill output. This creates an intent/documentation mismatch about what the skill actually includes in the free version.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.