Interrogate
Security checks across malware telemetry and agentic risk
Overview
This is a simple clarification-question skill that only changes how the agent asks follow-up questions and does not request system access or hidden execution.
This skill appears safe to install as a guided questioning aid. Users should expect it to activate only when they type /interrogate and to ask clarifying questions before proceeding. The listed crypto/purchase capability tags appear inaccurate, so rely on the actual artifact behavior rather than those tags.
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.