The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it directly interpolates untrusted user inputs into the prompts used for image generation.
Ingestion points: User inputs brand_name, industry, personality, and color_preference are gathered in SKILL.md.
Boundary markers: There are no delimiters or 'ignore embedded instructions' warnings around the interpolated variables in the prompt templates.
Capability inventory: The skill uses the muapi image generate tool and curl for network requests to generate assets.
Sanitization: No escaping, validation, or filtering of the user-provided text is performed before it is sent to the image generation models.
[COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands, specifically the muapi CLI and curl for API interactions. This includes using the MUAPI_API_KEY environment variable to authenticate requests to the vendor's API endpoint (api.muapi.ai).

muapi-brand-kit