ClawHub
Back to skill

Security audit

Strategic Thinking & Mental Models Engine

Security checks for vulnerabilities and agentic risk

Overview

This is a markdown-only decision-making framework skill with broad activation phrases but no code execution, persistence, credential handling, or hidden data movement.

Installers should expect this skill to appear for common decision-making prompts. Use it for structured analysis, but treat financial, legal, hiring, or other consequential outputs as advisory and review them with appropriate human judgment.

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

Medium
Confidence
96% confidence
Finding
The invocation phrase "Help me decide [your decision]" is extremely broad and overlaps with ordinary user language, making accidental or opportunistic triggering likely in unrelated conversations. Because this skill can activate on many normal decision-oriented prompts, it may hijack routing, inject unsolicited guidance, or interfere with other skills and system behaviors.

Vague Triggers

Medium
Confidence
97% confidence
Finding
The trigger phrase "help me decide" / "analyze this decision" is broad enough that the skill may activate in many ordinary conversations, including contexts where the user did not explicitly intend to invoke this skill. This can cause unintended routing or prompt hijacking at the orchestration layer, though the skill itself contains no directly dangerous actions or code execution behavior.

Static analysis

No suspicious patterns detected.