dotnet-new-app-slnx
Pass
Audited by Gen Agent Trust Hub on Apr 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a bundled PowerShell script (
scripts/resolve-package-versions.ps1) to resolve the latest stable package versions for the generated projects. - [EXTERNAL_DOWNLOADS]: The skill and its scripts fetch configuration and metadata from well-known services, including
api.nuget.organdraw.githubusercontent.com(specifically for Microsoft .NET release notes). - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to the handling of external data. * Ingestion points: The skill ingests user input via
FORMS.mdand external metadata from the NuGet API and GitHub release notes. * Boundary markers: The template assets do not use specific delimiters or instructions to prevent the agent from interpreting ingested strings (such as solution names or version strings) as commands. * Capability inventory: The skill can execute local PowerShell scripts and perform extensive file system writes during scaffolding. * Sanitization: No explicit validation or escaping is applied to external content or user parameters before they are interpolated into the solution's project files and build configurations.
Audit Metadata