Amazon Brand Accelerator
Security checks across malware telemetry and agentic risk
Overview
The skill bundle is mostly coherent, but one review helper defaults to launching a nested Codex reviewer with approval bypass and full sandbox access.
Install only if you trust the publisher and need these ClawHub maintainer workflows. Before using `autoreview`, consider running it with `--no-yolo` or `AUTOREVIEW_YOLO=0` so the nested reviewer does not bypass approvals and sandboxing by default. Treat moderation, production Convex, and PR-publishing commands as high-impact actions and keep the existing confirmation and dry-run steps.
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
65/65 vendors flagged this skill as clean.