The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local system commands for environment detection and validation purposes. Specifically, it uses grep for idempotency checks, dotnet build or python -m py_compile for syntax validation, and curl for local preflight request testing against localhost.
[EXTERNAL_DOWNLOADS]: The diagram.html file includes a reference to the Mermaid.js library via cdn.jsdelivr.net. This is a well-known and trusted Content Delivery Network (CDN) service used for rendering workflow diagrams.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an indirect injection surface as it ingests data from untrusted local project files.
Ingestion points: Reads project directory structures and configuration files like .env and appsettings.json to detect frontend origins.
Boundary markers: No explicit delimiters or instructions to ignore embedded content are defined for the input files.
Capability inventory: The skill can perform file writes, execute build commands, and trigger local network requests via curl.
Sanitization: The skill does not explicitly detail sanitization logic for the data read from project files before using it in code generation.

ln-773-cors-configurator