ClawHub
Back to skill

Security audit

Resistance Training Equipment Buying Consultant

Security checks for vulnerabilities and agentic risk

Overview

This is a plain instruction-only buying guide for home strength equipment, with a minor wording mismatch around brand-neutral advice but no hidden execution, persistence, credential use, or unsafe system access.

Use this skill for structured equipment-buying guidance, but treat any named products as examples to verify independently for current specs, price, availability, warranty, and fit. Avoid sharing an exact home address; an approximate region is enough. For heavy equipment on upstairs or uncertain floors, confirm loading limits with a qualified professional.

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
  • 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
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Intent-Code Divergence

Medium
Confidence
97% confidence
Finding
The instructions explicitly say the goal is not to recommend specific brands, but the workflow later requires suggesting branded products. Such contradictory guidance can be exploited to justify brand steering while preserving a false appearance of neutrality, which undermines trust and can produce inconsistent agent behavior.

Description-Behavior Mismatch

Medium
Confidence
95% confidence
Finding
The skill presents itself as brand-neutral, but later instructs the agent to recommend specific real products and named models. This creates an internal policy contradiction that can mislead users about impartiality and open the door to biased or undisclosed promotional behavior, even if no direct affiliate mechanism is present.

Static analysis

No suspicious patterns detected.