The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The instructions utilize uvx and npx to dynamically fetch and run datamodel-code-generator and json-schema-to-typescript from their respective public package registries.
[COMMAND_EXECUTION]: The skill executes several shell commands to manage the SDK generation process, including running schema validation scripts via mise and invoking the code generators with specific configuration flags.
[DYNAMIC_EXECUTION]: After generating the SDK artifacts, the skill executes a verification step that attempts to import the newly created Python models using python3 -c to ensure they are syntactically valid and loadable.
[INDIRECT_PROMPT_INJECTION]: The skill processes JSON schema files located in the schemas/bloodbank/ directory to produce code.
Ingestion points: JSON files matching schemas/bloodbank/**/*.json are used as input for the generators.
Boundary markers: No specific delimiters or warnings are used when processing the schema content.
Capability inventory: The skill environment allows for shell command execution, filesystem writes, and the use of package runners.
Sanitization: The skill assumes the integrity of the local schema files and relies on the underlying generation tools to parse the schema data correctly.

bloodbank-sdk-generation