Disease Investigation
Security checks across malware telemetry and agentic risk
Overview
The skill artifacts are mostly purpose-aligned, but one review helper defaults to launching a nested agent with full local access and disabled approvals.
Install only if you trust these workflows in the target repo. For autoreview, consider using `--no-yolo` or `AUTOREVIEW_YOLO=0` by default, and be aware that fallback reviewers may receive generated diffs. For moderation and PR-maintainer use, confirm targets and reasons before writes because those commands can affect users, public content, and repository state.
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.