The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the muapi CLI tool and curl commands to interact with the api.muapi.ai endpoint. These commands include muapi image generate, muapi image edit, muapi auth configure, and muapi predict wait to manage image processing tasks.
[INDIRECT_PROMPT_INJECTION]: The skill ingests user-supplied text to construct prompts for image generation and layout planning, creating a potential surface for indirect injection if inputs contain malicious instructions.
Ingestion points: User inputs title, channel_style, and subject_description are processed in Phase B and Phase C of SKILL.md.
Boundary markers: The skill lacks explicit boundary markers or instructions for the model to ignore potential commands embedded within the user-provided strings.
Capability inventory: The skill has the capability to execute shell commands (curl) and specialized CLI tools (muapi) based on instructions derived from the input data.
Sanitization: There are no defined filters, escaping mechanisms, or validation steps applied to the user inputs before they are interpolated into the workflow.

muapi-youtube-thumbnail