AANA Continuous Self-Improvement Skill
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only self-review skill that tells agents to improve workflows while preserving user approval, privacy, and scope boundaries.
Reasonable to install for structured self-review. Review any future request from the agent that would save memory, change files, alter tools or policies, add automation, or send a review payload to an external checker; keep payloads redacted and ignore the unrelated marketplace crypto/purchase tags unless the publisher corrects them.
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
64/64 vendors flagged this skill as clean.