debate-research
Security checks across malware telemetry and agentic risk
Overview
This skill transparently orchestrates debate subagents and web search to produce a research report, with no hidden install, persistence, or unrelated data access found.
Install if you are comfortable with your debate topic and generated arguments being processed by configured LLM providers and with Phase 1 using web search. Keep plan confirmation enabled for cost and scope control, avoid highly sensitive topics, and only set output_path to a location where you intend a report to be written.
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
61/61 vendors flagged this skill as clean.