The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the processing of potentially untrusted data through ingestion points such as EditForm components, JSInvokable methods, and parameter binding. While boundary markers like explicit delimiters are not defined in the instructions, the skill mitigates risk through its capability inventory which is scoped to developer tasks like editing files and running builds, and by providing sanitization patterns using DataAnnotationsValidator and server-side authorization checks.
[DYNAMIC_EXECUTION]: The skill includes patterns for DynamicComponent and JavaScript interop via IJSRuntime. These are core components of the Blazor framework used for runtime UI composition and browser integration, presented here with standard implementation examples.
[COMMAND_EXECUTION]: The skill description specifies that the agent will perform tasks such as inspecting the repository, editing files, and running build, test, or lint commands. These capabilities are directed toward project maintenance and validation within the scope of a C# Blazor environment.
[SAFE]: The inclusion of a detailed 'Security Anti-Patterns' section (references/anti-patterns.md) actively guides the agent to avoid common security vulnerabilities, such as exposing API keys in component state or relying exclusively on client-side security checks.

blazor