The Agent Skills Directory

[DATA_EXFILTRATION]: The skill performs network operations by sending user-provided image URLs and an API key to the api.muapi.ai domain via the muapi CLI or curl. While these operations are necessary for the skill's primary function, the destination domain is not a whitelisted source for data transfer.
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external image data which is used to influence the AI's generation of stickers, poses, and captions, creating a potential surface for indirect prompt injection.
Ingestion points: Untrusted data enters the agent context through the person_image URL provided by the user, as defined in SKILL.md.
Boundary markers: No explicit delimiters or instructional warnings are present to separate the external content from the generation prompt.
Capability inventory: The skill utilizes muapi CLI tools (muapi image edit, muapi predict wait) and curl for network requests to external image processing endpoints.
Sanitization: The skill performs no local validation or sanitization of the input URL; it relies entirely on the external API's internal moderation filter, which is explicitly configured to a low level.

muapi-chibi-collage-effect