The Agent Skills Directory

[COMMAND_EXECUTION]: The skill includes a local Python script scripts/apimart_batch_generate.py which is designed to be executed by the agent to automate the image generation process using the APIMart API.
Evidence: The execution flow is described in stages/04-prompts.md and scripts/README.md, which provide command-line instructions for running the batch generator.
[EXTERNAL_DOWNLOADS]: The provided batch generation script makes outbound network requests to api.apimart.ai to send prompts and download generated image assets.
Evidence: The post_json and download_url functions in scripts/apimart_batch_generate.py interact with the external API to transmit data and retrieve binary image content.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided articles to generate image prompts, creating a surface for potential injection attacks.
Ingestion points: User-supplied text content provided during the demand clarification phase in stages/01-brief.md.
Boundary markers: The skill uses template placeholders (e.g., {标题}, {结论}) in templates/ but lacks robust sanitization or specific delimiters to isolate user-provided content from instructions.
Capability inventory: The agent can execute a local script capable of network communication and writing files to the local disk.
Sanitization: There is no evidence of filtering or escaping logic applied to user content before it is interpolated into the prompts for the image generation model.

image-assistant