Chinese Medicine
Security checks across malware telemetry and agentic risk
Overview
The reviewed skill artifacts are coherent repo-maintenance and Convex helper workflows, with sensitive commands disclosed and mostly gated by user intent or explicit confirmation.
Install only if you want agent help with ClawHub and Convex maintenance. Before running moderation, proof publishing, Convex deployment, or autoreview commands, confirm the target repository/account and use dry-run or opt-out flags where appropriate, especially `--no-yolo` for autoreview if you do not want broad nested Codex access.
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
64/64 vendors flagged this skill as clean.