winapp-frameworks

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • External Package Installation: The skill instructs the agent to install @microsoft/winappcli via npm and Microsoft.winappcli via Winget. These packages originate from the vendor's official infrastructure and are used to provide the development tools described.
  • External Resource References: The skill contains numerous links to documentation and guides hosted on the vendor's official GitHub repository (github.com/microsoft/WinAppCli). These links provide supplementary information for framework setup and packaging.
  • Command Execution: The instructions include several command-line examples for building, initializing, and running applications (e.g., dotnet build, winapp init, winapp run). These commands are standard for the developer workflows supported by this skill.
  • Indirect Prompt Injection Surface: The skill processes project-level configuration data from files such as .csproj, package.json, and Package.appxmanifest. While these files are integral to the build process, they represent a potential surface where untrusted content in a project could influence agent behavior.
  • Ingestion points: Project configuration files such as .csproj, package.json, and Package.appxmanifest (SKILL.md).
  • Boundary markers: No explicit boundary markers or "ignore" instructions for embedded data are defined.
  • Capability inventory: The skill utilizes tool execution capabilities for building and running code across multiple development frameworks.
  • Sanitization: No specific sanitization or validation of the input configuration files is mentioned in the prompt instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 02:30 PM
Security Audit — agent-trust-hub — winapp-frameworks