skills/ingpoc/skills/winui-app/Gen Agent Trust Hub

winui-app

Warn

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill automates environment preparation by executing winget configure -f config.yaml. This command applies a Desired State Configuration (DSC) that performs high-risk system modifications.
  • [COMMAND_EXECUTION]: The skill performs privilege escalation by requesting an elevated security context in config.yaml to enable Windows Developer Mode and install system-level software (Visual Studio Community).
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection during the application scaffolding process. It takes a user-provided application name and interpolates it directly into shell commands without rigorous sanitization.
  • Ingestion points: User-supplied project names captured during the Required Flow in SKILL.md.
  • Boundary markers: Absent. The user input is directly concatenated into strings for dotnet new and dotnet build commands.
  • Capability inventory: The agent has access to dotnet CLI tools and winget for shell-based project creation and building.
  • Sanitization: Absent. Although the instructions mention using a "safe folder name," there is no defined validation logic to prevent command injection characters (e.g., semicolons or pipes) from being included in the name.
  • [EXTERNAL_DOWNLOADS]: The skill fetches software and components from Microsoft's official repositories. It attempts to install "Visual Studio Community 2026" and specific workloads using the Microsoft.VisualStudio.Community package ID. While these sources are trusted, the automated installation of remote binaries with administrative privileges is a notable security surface.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jun 15, 2026, 09:46 AM
Security Audit — agent-trust-hub — winui-app