ClawHub
Back to skill

Security audit

think-like-fable

Security checks for vulnerabilities and agentic risk

Overview

This is a reasoning-style skill that broadens when an agent should be careful, but it does not request files, credentials, network access, persistence, or execution authority.

Installers should understand that this skill may activate on broad phrases like “be rigorous” or “are you sure?” and may influence the agent’s reasoning style in high-stakes tasks. That is disclosed in the artifact and does not appear to grant extra system access.

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
Confidence
91% confidence
Finding
The trigger list is broad and includes common phrases such as "be rigorous," "think hard about this," and "are you sure?" that can appear in ordinary conversation. This can cause unintended skill activation, altering the agent's behavior unexpectedly and increasing prompt-surface exposure in contexts where this meta-skill was not intended to run.

Vague Triggers

Medium
Confidence
88% confidence
Finding
The usage section allows activation based on subjective conditions like "high-stakes decision," "security judgment," or when another prompt says to apply the manual. These criteria are insufficiently constrained, making activation dependent on fuzzy interpretation and allowing other prompts or routine tasks to pull this skill in unexpectedly, which can create unpredictable behavior and prompt-injection-style control over execution flow.

Static analysis

No suspicious patterns detected.