AANA Evidence First Answering Skill
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only skill for evidence-checking answers, with no code execution or persistence and clear privacy limits.
Install this as a process aid, not as proof that an answer is correct. When it suggests retrieval or an AANA checker, use only user-approved tools and keep secrets, raw records, full logs, private messages, and unrelated private data out of review payloads.
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.