AANA Task Scope Guardrail Skill
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only scope guardrail that tells agents to use less data, ask before expanding work, and stop when the task is complete.
This skill is reasonable to install if you want agents to stay tightly within user-requested task boundaries. Expect it to make the agent ask before broader work and avoid unrelated private data; the publisher should fix the unrelated marketplace tags so users are not misled about its purpose.
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
62/62 vendors flagged this skill as clean.