Brain Evolution Skill

Security checks across malware telemetry and agentic risk

Overview

This appears to be a memory/persistence skill, but its privacy promises are inconsistent enough that users should review it before installing.

Review before installing. Treat any saved memories, caches, traces, or logs as potentially containing sensitive text unless you can verify redaction is enforced. Avoid enabling automatic recording, limit collection to non-sensitive inputs, and inspect or clear the storage location regularly.

SkillSpector

By NVIDIA

Vulnerability Patterns

MCP Tool PoisoningHidden Instructions, Unicode Deception, Parameter Description Injection
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)

Intent-Code Divergence

Medium

Confidence: 95% confidence
Finding: The documentation claims that sensitive conversations will not be recorded, but elsewhere states that sensitive-word filtering is only a future improvement. This creates a misleading privacy guarantee: users or integrators may rely on non-existent protections and inadvertently store sensitive prompts, tool outputs, or personal data in the skill's memory, cache, trace logs, or SQLite/JSON persistence layers.

VirusTotal

63/63 vendors flagged this skill as clean.

View on VirusTotal