Deep Planning Skill
Security checks across malware telemetry and agentic risk
Overview
This is a planning-only skill that uses an isolated subagent for complex design work and does not include executable code, hidden data access, or persistence.
This skill appears safe to install for users who want a structured planning workflow. Be aware that it can ask the agent to use an isolated subagent for complex planning, which may add latency and token cost, but the artifact explicitly limits what that subagent should receive.
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.