ClawHub

Enhanced Agent Modes

Security checks across malware telemetry and agentic risk

Overview

The skill is documentation-only and not malicious, but it asks agents to support memory persistence and optional unattended background work without enough scope or control details.

Install only if you want agent behavior controls beyond basic modes. Keep auto_memory and autonomous_crons disabled unless you explicitly need them, and require confirmation before any memory scan, long-term memory update, note archive, scheduled task, or spawned sub-agent run.

SkillSpector

By NVIDIA
Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Missing User Warnings

Medium
Confidence
93% confidence
Finding
The skill describes an automatic memory-consolidation flow that scans daily memory files, updates long-term memory, and archives notes, but it does not require explicit user consent, clear disclosure of what data is processed, or safeguards around what files are eligible. In an agent environment, this can cause unintended collection, retention, or modification of sensitive user/project data and may normalize autonomous state-changing behavior that the user did not knowingly authorize.

Missing User Warnings

Medium
Confidence
95% confidence
Finding
The skill documents autonomous background tasks and spawning sub-agents on timer events without prominent user warning, approval gates, or strict action boundaries. Asynchronous execution is especially risky in agent skills because it can trigger actions outside the user's immediate awareness, potentially causing unintended tool use, external requests, data access, or persistent changes.

VirusTotal

64/64 vendors flagged this skill as clean.

View on VirusTotal