The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions in SKILL.md direct the agent to execute shell commands using the muapi CLI tool (e.g., muapi auth configure, muapi image generate) and curl to interact with the service endpoint.
[EXTERNAL_DOWNLOADS]: The skill interacts with the external domain api.muapi.ai and requires the muapi CLI utility to be present in the environment to perform image generation and editing tasks.
[INDIRECT_PROMPT_INJECTION]: The skill lacks sanitization for user-controlled inputs interpolated into image generation prompts.
Ingestion points: Untrusted data enters via the topic, style, and text_overlay input variables in SKILL.md.
Boundary markers: No delimiters or specific instructions are used to prevent the agent from misinterpreting user input as instructions.
Capability inventory: The skill has the capability to execute CLI commands and perform network operations via curl as defined in SKILL.md.
Sanitization: No escaping or validation is performed on the user inputs before they are included in the prompt strings.

muapi-rednote-cover