humanizer-academic-en
Security checks across malware telemetry and agentic risk
Overview
The skill collection mostly does what it says, but one review helper can run a powerful nested code-review command with safeguards disabled by default.
Install only in a trusted ClawHub maintainer environment. Before using `autoreview`, consider opting out of full-access mode with `--no-yolo` or `AUTOREVIEW_YOLO=0`, and disable fallback reviewers if diffs should not be sent to other local AI CLIs. Treat moderation and PR-publishing workflows as privileged actions that need explicit targets, reasons, and confirmation.
SkillSpector
By NVIDIA
Vulnerability Patterns
- 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
- Excessive AgencyUnrestricted Tool Access, Autonomous Decision Making, Scope Creep
VirusTotal
63/63 vendors flagged this skill as clean.