Pydantic Ai Dependency Injection
Security checks across malware telemetry and agentic risk
Overview
This is a documentation-only PydanticAI dependency-injection skill with no executable code, though users should carefully scope any database or credential examples they adapt.
Safe to install as documentation. When adapting the examples, avoid arbitrary database query tools, use parameterized or purpose-specific operations, pass initialized clients instead of raw secrets where possible, and keep any API keys out of source code and logs.
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
67/67 vendors flagged this skill as clean.