ClawHub
Back to skill

Security audit

Chain Reason

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only reasoning-format skill with no code, external access, persistence, or hidden data handling, though its activation wording is broad.

Install this only if you want the agent to use structured reasoning-style formats frequently. Be aware that common prompts asking how or why may trigger it, which can make answers longer or more procedural than usual.

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

High
Confidence
97% confidence
Finding
The activation criteria are extremely broad and keyed to common words like "explain," "how," "why," and "reason," which are likely to appear in many ordinary prompts. That can cause the skill to activate far more often than intended, changing model behavior globally and increasing exposure of internal reasoning-style traces or policy-driven decision scaffolding in benign conversations.

Vague Triggers

Medium
Confidence
89% confidence
Finding
The usage guidance relies on subjective conditions like "tradeoffs," "multiple ways," and "requires judgment" without operational thresholds. In practice, this ambiguity makes activation inconsistent and overbroad, which can lead to unnecessary disclosure of reasoning traces, prompt-surface expansion, and interference with safer default response behavior.

Static analysis

No suspicious patterns detected.