The Agent Skills Directory

[COMMAND_EXECUTION]: The core workflow requires the agent to execute shell commands including dotnet build and dotnet test to verify application integrity, as well as curl for manual endpoint testing.- [INDIRECT_PROMPT_INJECTION]: The skill defines several API ingestion points (e.g., CreateProductRequest and RegisterCommand) that process external data. If the agent later processes outputs or logs from these services, it could be influenced by embedded instructions. Ingestion points: MediatR command handlers and Minimal API route handlers. Boundary markers: None explicitly defined in the instructions for data processing. Capability inventory: The agent can write files and execute shell commands. Sanitization: The skill implements FluentValidation as a defense-in-depth measure.- [DYNAMIC_EXECUTION]: The testing workflow uses dotnet test, which involves the execution of code compiled at runtime from source files generated or modified during the task.

dotnet-core-expert