ClawHub
Back to skill

Security audit

Ruthless Mentor

Security checks for vulnerabilities and agentic risk

Overview

This is a simple review-helper skill that delegates plan critique to a sub-agent, with minor scope concerns around broad trigger wording.

Install only if you want an intentionally critical planning reviewer. Use it with explicit prompts and avoid sending confidential plans unless you are comfortable with them being evaluated by the configured sub-agent/model; enable any weekly review or workflow gate only after clear confirmation.

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
Category
Not specified by scanner
Confidence
90% confidence
Finding

The trigger guidance includes broad language such as using the skill whenever someone asks 'should we do X?', which can overlap with ordinary planning conversations and cause the skill to be invoked outside clearly bounded contexts. Because this skill spawns a sub-agent and applies a strong adversarial evaluation prompt, unintended activation could alter workflow decisions or introduce unnecessary tool use during normal discussion.

Content

No source excerpt is available for this finding.

Vague Triggers

Medium
Category
Not specified by scanner
Confidence
95% confidence
Finding

The manual trigger phrase 'run this through the mentor' is generic and easy to collide with normal conversational phrasing, making accidental or adversarial activation more likely. In this skill, activation leads to delegated analysis through sessions_spawn, so a loose trigger increases the chance of unintended sub-agent execution and decision-shaping behavior.

Content

No source excerpt is available for this finding.

Static analysis

No suspicious patterns detected.