The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill interpolates user-supplied text directly into instructions provided to downstream image generation models.
Ingestion points: User inputs for product_name, background_style, lighting, and category defined in SKILL.md.
Boundary markers: Absent; user strings are concatenated directly into the tool prompts without specific delimiters or instructions to ignore embedded commands.
Capability inventory: The skill utilizes muapi CLI tools and curl to perform network-based image generation tasks.
Sanitization: No sanitization, validation, or escaping of the user-provided strings is implemented in the skill instructions.
[COMMAND_EXECUTION]: The skill requires the executing agent to run shell commands to interact with the MuAPI platform.
Evidence: The instructions specify the use of muapi image generate and muapi auth configure commands.
[DATA_EXFILTRATION]: The skill performs network operations to the MuAPI service infrastructure.
Evidence: The skill makes network requests to api.muapi.ai using curl and the muapi CLI tool to transmit generation parameters and retrieve results.

muapi-multi-angle-shots