ClawHub
Back to skill

Security audit

ai-brain-learning-memory

Security checks for vulnerabilities and agentic risk

Overview

This is a low-risk informational AI memory methodology skill with one broad-trigger routing concern but no hidden access, network use, credential handling, or unsafe persistence.

Before installing, be aware that the skill may activate on general conversations about AI, learning, or memory. It does not appear to access private data or run network operations, but users who want precise routing should narrow its trigger phrases.

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 (1)

Vague Triggers

Medium
Confidence
86% confidence
Finding
The manifest advertises many generic trigger phrases such as memory, learning, forgetting, and related English variants, which can cause the skill to activate on broad everyday conversations outside the user’s likely intent. In a memory-related guidance skill, this creates overreach and prompt-routing risk rather than direct code execution risk, but it can still lead to unintended invocation, user confusion, and inappropriate influence over unrelated discussions.

Static analysis

No suspicious patterns detected.