The Agent Skills Directory

[CREDENTIALS_UNSAFE]: The skill's instructions in SKILL.md direct the agent to handle and transmit a secret MUAPI_API_KEY. It specifically instructs the agent to include this key in the headers of curl requests sent to a remote API endpoint.
[COMMAND_EXECUTION]: The skill requires the execution of shell commands, including the muapi CLI and curl, to perform its primary functions as described in the 'Notes for the Executing Agent' section of SKILL.md.
[EXTERNAL_DOWNLOADS]: The skill initiates network requests to api.muapi.ai via curl. This domain is not recognized as a whitelisted or well-known trusted service.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill accepts untrusted data from the user via the url, page_type, and redesign_style input fields in SKILL.md.
Boundary markers: No delimiters or protective instructions are used to isolate user-provided text from the instructional prompts.
Capability inventory: The skill has the capability to execute shell commands and perform network operations using curl and the muapi CLI.
Sanitization: The instructions do not include any mechanisms for sanitizing or validating user input before it is interpolated into the prompts for image generation.

muapi-url-to-design