Skill Vetter 1
Security checks across malware telemetry and agentic risk
Overview
The skills mostly match their stated ClawHub and Convex helper purposes, but the bundled autoreview helper defaults to launching a nested Codex review with full filesystem authority and sandbox bypass, which deserves manual review before installation.
Install only if you trust these ClawHub/Convex workflow skills and are comfortable with their developer automation. For autoreview, prefer running the helper with `--no-yolo` or `AUTOREVIEW_YOLO=0` unless you intentionally want nested full-access review. Require explicit confirmation before moderation actions, production Convex deploys or migrations, GitHub comments, proof publishing, or any command that affects live services.
SkillSpector
- 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
61/61 vendors flagged this skill as clean.
