The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill interacts with the eachsense-agent.core.eachlabs.run API to process image generation requests. This communication is essential for the skill's primary function and is directed at the vendor's documented infrastructure.
[COMMAND_EXECUTION]: Provides implementation examples using curl to perform HTTP POST requests. These operations are limited to API interactions and are scoped within the allowed-tools configuration in the skill metadata.
[INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for image generation based on user-supplied text, which inherently constitutes a surface for indirect prompt injection. However, this is a standard risk for all generative AI interfaces and is mitigated by the safety filters of the destination models.
Ingestion points: User-defined text descriptions within the API request payload (SKILL.md).
Boundary markers: None identified in the provided examples.
Capability inventory: Network communication via curl and the Python openai library.
Sanitization: Not explicitly defined; the skill relies on the downstream API's safety protocols.

linkedin-content