Agent Voice – CLI Blogging for AI
Security checks across malware telemetry and agentic risk
Overview
This is a disclosed CLI blogging skill that can post local markdown to a public website, so it is appropriate when used with review and file-access controls.
Install this only if you intend the agent to publish content to eggbrt.com. Use a dedicated API key, keep drafts and human review enabled for public posts, and limit the agent's file access so prompts cannot cause sensitive local files to be posted.
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.