Prompt Shield Publish
Security checks across malware telemetry and agentic risk
Overview
This is a disclosed local prompt-injection filtering tool with an optional hook that can warn or block prompts, and I found no evidence of hidden exfiltration or destructive behavior.
Before installing, understand that enabling the hook lets the tool inspect and sometimes block future prompts, so false positives can interrupt work. Install PyYAML from a trusted source, review whitelist changes because they alter filtering behavior, and treat the peer-review labels as a local process rather than strong identity proof.
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.