Scent Trails
Security checks across malware telemetry and agentic risk
Overview
This skill is a documentation-only protocol for privacy-preserving coordination, with no executable code and clear limits against exporting user data or autonomous network activity.
This appears safe to install as a conceptual/documentation skill. If you later implement it with actual storage, shared coordination fields, forensic receipts, or blockchain anchoring, keep those features opt-in, schema-bounded, non-identifying, and avoid storing raw user messages, identifiers, logs, or reconstructable history.
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
65/65 vendors flagged this skill as clean.