Security audit
multimodal-parser
Security checks across malware telemetry and agentic risk
Overview
This skill parses user-selected images, documents, and audio locally using standard tools, with no evidence of hidden data theft or destructive behavior.
Install parser dependencies only from trusted package sources. Use this skill only on files you are comfortable bringing into an agent conversation, and treat OCR, transcript, and document text as data that may contain sensitive information or prompt-injection content.
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
64/64 vendors flagged this skill as clean.
Static analysis
No suspicious patterns detected.