The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Node.js scripts (generate.mjs and build_catalog.mjs) to interact with the apimodels.app API and manage the local model catalog.
[EXTERNAL_DOWNLOADS]: The generate.mjs script includes functionality to download generated images from the service's API and save them to a user-defined local path via the --out argument.
[PERSISTENCE]: The skill provides instructions for persisting the API key by adding it to the user's shell profile (~/.zshrc) or saving it to a dedicated credentials file (~/.apimodels/credentials). The script automatically sets restricted file permissions (0600) on this credentials file to prevent unauthorized access.
[INDIRECT_PROMPT_INJECTION]: The skill acts as a proxy for user-generated prompts and remote image URLs, which are processed by third-party image models.
Ingestion points: User-provided text prompts and reference image URLs passed as arguments to the generate.mjs script (found in SKILL.md).
Boundary markers: The skill relies on the service provider's built-in content moderation filters, which are explicitly referenced in the error handling logic.
Capability inventory: The skill possesses network access to the apimodels.app domain and the ability to write files to the local filesystem.
Sanitization: Implements robust secret management practices, including reading keys from stdin to avoid shell history and masking sensitive values in console output.

apimodels-image