fitness-functions
Pass
Audited by Gen Agent Trust Hub on Mar 18, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data access behaviors were detected across the skill files and templates.
- [COMMAND_EXECUTION]: The skill utilizes standard .NET CLI commands (dotnet new, dotnet add) which are legitimate and necessary for setting up architecture test projects.
- [EXTERNAL_DOWNLOADS]: Dependencies are restricted to reputable, community-standard packages (NetArchTest.Rules and ArchUnitNET) from the official NuGet registry.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it analyzes untrusted local source code and project structures. Documentation: (1) Ingestion points: local solution, project files, and namespace hierarchies analyzed in the workflow. (2) Boundary markers: Not explicitly defined. (3) Capability inventory: Skill tools include Read, Write, Bash, Task, and Skill. (4) Sanitization: None observed, as the skill functions as a static analysis and reporting tool.
Audit Metadata