Concept Decoder
Security checks across malware telemetry and agentic risk
Overview
The skills are mostly coherent developer and staff workflows, but they include high-impact moderation authority and a review helper that runs nested Codex with full sandbox bypass by default.
Install only if you trust this publisher and intend to use it for ClawHub maintainer work. Before running autoreview, prefer `--no-yolo` unless full local access is acceptable. Use staff moderation commands only with explicit targets, reasons, confirmation, and an account that is allowed to perform those actions.
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.