Clear-Mind Skill (龙虾大脑自动清理技能)

• Outcome: Review
• Latest audit: May 29, 2026 at 9:24 PM UTC
• Version:
  1.0.2

Overview

This memory-management skill appears purpose-aligned, but it can restructure or delete user memory with unclear approval and rollback safeguards.

Install only if you are comfortable with the agent modifying long-term memory files. Before using it, require a dry-run summary, a timestamped backup, explicit approval for deletions or migrations, and a simple restore procedure.

SkillSpector

By NVIDIA

Vulnerability Patterns

• Data Exfiltration: External Transmission, Env Variable Harvesting, File System Enumeration
• Trigger Abuse: Overly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
• Prompt Injection: Instruction Override, Hidden Instructions, Exfiltration Commands
• Privilege Escalation: Excessive Permissions, Sudo/Root Execution, Credential Access
• Supply Chain: Unpinned Dependencies, External Script Fetching, Obfuscated Code

Findings (2)

Vague Triggers

Medium

• Confidence: 87% confidence
• Finding: The activation criteria are broad enough that the skill may be invoked for routine or ambiguous situations such as generic 'memory cleanup' or periodic maintenance, even when no clear user-approved need exists. Because the skill performs restructuring, migration, and deletion of user data, underspecified triggers increase the chance of unnecessary modification or loss of important information.

Missing User Warnings

Medium

• Confidence: 91% confidence
• Finding: The skill authorizes moving content, deleting redundant or outdated notes, and creating new files, but the safety warning and consent requirements are not enforced at each destructive step. In a memory-management context, this is dangerous because an agent could modify or remove user data based on imperfect classification, causing silent data loss or corruption of long-term memory structure.

VirusTotal

63/63 vendors flagged this skill as clean.