DCL Semantic Drift Guard — Hallucination & Context Drift Detector
Security checks across malware telemetry and agentic risk
Overview
This is an instruction-only skill for checking AI output against supplied sources, with no hidden install steps or privileged behavior found.
Install if you want a prompt workflow for checking generated text against trusted sources. Use care with confidential documents and only provide kb_endpoint URLs you trust; the artifact does not independently prove the publisher's no-retention, compliance, or audit-chain claims.
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
66/66 vendors flagged this skill as clean.