ClawHub

LYGO Universal Living Memory Library (v1.1)

Security checks across malware telemetry and agentic risk

Overview

This skill is mostly a local memory-audit helper, but it tells users to publicly anchor archives made from logs and conversation exports without enough privacy safeguards.

Install only if you will treat public anchoring as opt-in. Before minting or anchoring any archive, remove secrets, personal data, internal links, identifiers, and private decisions; use local or private verification for sensitive material.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • MCP Least PrivilegeUnderdeclared Capability, Wildcard Permission, Missing Permission Declaration
  • 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 (2)

Lp3

Medium
Category
MCP Least Privilege
Confidence
84% confidence
Finding
The skill advertises itself as a 'pure advisor' and does not declare permissions, but it explicitly references scripts that run audits and mentions capabilities consistent with reading and writing files. That mismatch can mislead users and host systems about what the skill may do, weakening least-privilege controls and making unintended file access harder to detect or govern.

Missing User Warnings

Medium
Confidence
94% confidence
Finding
The protocol explicitly instructs users to 'anchor it publicly' after minting a master archive derived from logs, conversation exports, scrolls, and decision receipts. Because those inputs can contain sensitive personal, operational, or confidential data, publishing the archive or its metadata without privacy guidance creates a real risk of unintended disclosure and permanent exposure.

VirusTotal

65/65 vendors flagged this skill as clean.

View on VirusTotal