Oc Extract Memories

• Outcome: Pass
• Latest audit: May 29, 2026 at 8:41 PM UTC
• Version:
  1.0.0

Overview

This skill is a disclosed local memory-saving helper; the main risk is that it may save chat details automatically when a conversation ends.

Install this only if you want your agent to save selected conversation details into local memory files. Avoid using it in chats containing secrets, credentials, or sensitive personal information unless you are comfortable reviewing and deleting generated entries in memory/topics/ and MEMORY.md.

SkillSpector

By NVIDIA

Vulnerability Patterns

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

Findings (2)

Vague Triggers

Medium

• Confidence: 94% confidence
• Finding: The skill auto-triggers on very common conversation-ending phrases and then performs persistence actions by analyzing the conversation and writing files. This can cause unintended activation, leading to storage of conversation-derived data without an explicit user request or fresh consent, which is risky for privacy and can create incorrect or sensitive long-term memory entries.

Missing User Warnings

Medium

• Confidence: 96% confidence
• Finding: The skill instructs the agent to write to memory/topics/ and update MEMORY.md automatically, but does not require a user-facing warning, preview, or confirmation. Automatic file writes based on conversational content can persist sensitive, mistaken, or unwanted data, especially when combined with the broad end-of-conversation auto-trigger behavior.

VirusTotal

66/66 vendors flagged this skill as clean.