ClawHub
Back to skill

Security audit

Personal Growth Coach

Security checks for vulnerabilities and agentic risk

Overview

This is a simple coaching skill that creates thinking exercises and keeps scoped learning notes, with no code execution or external data transfer.

Install this if you are comfortable with the agent keeping local notes about your quiz performance and weak areas for future sessions. Review or delete the `memory/personal-growth-records/` notes if you do not want that history retained.

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
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
Findings (3)

Vague Triggers

Medium
Confidence
96% confidence
Finding
The trigger phrases and activation conditions are broad enough to capture ordinary requests about improving thinking or communication, which can cause the skill to activate when the user did not explicitly ask for this workflow. Because the skill then reads and updates persistent records, unintended invocation can lead to privacy-impacting behavior and unexpected state changes.

Autonomous Decision Making

Medium
Category
Excessive Agency
Content
### Quiz Principles
- ❌ No multiple choice or true/false questions
- ❌ No asking users to "recall a situation" or "remember an experience"
- ✅ Direct questions for users to answer
- ✅ General topics (work efficiency, time management, communication)
- ✅ Avoid overly specialized/industry-specific scenarios
Confidence
75% confidence
Finding
Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.

Missing User Warnings

Medium
Confidence
96% confidence
Finding
The skill documents that it reads prior learning records and updates session logs, but it does not clearly warn users in the description or trigger behavior that persistent memory will be accessed and modified. This reduces informed consent and can surprise users with retention of behavioral or performance data across sessions.

Static analysis

No suspicious patterns detected.