ClawHub
Back to skill

Security audit

Skilled Models Advisor

Security checks for vulnerabilities and agentic risk

Overview

This skill is a model-selection helper that delegates to a model database/agent and does not ship executable code or hidden persistence.

Before installing, expect the agent to consult the tool-models helper for model, provider, pricing, and capability questions. Review recommendations before acting on cost-sensitive or production configuration changes, since the skill optimizes around current access and provider data.

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 (1)

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
94% confidence
Finding

The description says to use the skill for 'any task' involving model choice and for broad questions like 'which model should I use for X?', without clearly bounding context or providing exclusion cases. This can cause the skill to trigger on ordinary conversational planning or generic advice requests beyond a narrowly defined invocation scope.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.