The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides instructional examples using curl to demonstrate how users can interact with the Each Labs API. These commands are for documentation purposes and are executed at the user's discretion.
[EXTERNAL_DOWNLOADS]: Includes examples that reference external image URLs (e.g., from example.com) to demonstrate multimodal capabilities, such as matching a visual style from a reference image.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided descriptions and images for content generation. While this represents a potential attack surface for indirect prompt injection, it is a standard feature of generative AI skills and is managed by the underlying model's safety protocols.
[SAFE]: No evidence of prompt injection, data exfiltration, obfuscation, or unauthorized privilege escalation was found. The skill uses environment variables for secret management and targets legitimate API endpoints.

pinterest-pins