Ai Voice Acting
Security checks across malware telemetry and agentic risk
Overview
The artifacts describe explicit ClawHub operations and review workflows with clear user-control gates and no evidence of hidden or deceptive behavior.
Install only if you expect ClawHub staff/admin and code-review workflows. The most sensitive actions are user bans, package or org changes, staff email, and production migrations, so use them with explicit targets, dry-run review, and signoff as the skill instructs.
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.