The Agent Skills Directory

[SAFE]: The skill uses official API endpoints for each::labs (eachlabs.run) for all image processing operations. All external URLs provided in the examples are either official documentation or standard placeholders (example.com).
[SAFE]: API keys are handled using environment variables ($EACHLABS_API_KEY) or standard placeholders (YOUR_EACHLABS_API_KEY), which follows secure credential management practices.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process natural language instructions and external image URLs to perform edits. While this creates an injection surface, it is the primary intended functionality of the skill.
Ingestion points: Processed via the messages array in the provided curl and Python examples in SKILL.md.
Boundary markers: Absent; instructions are passed directly as text to the AI model.
Capability inventory: The skill utilizes curl and the openai Python SDK to communicate with a remote API.
Sanitization: Not explicitly implemented in the provided documentation snippets.

image-editing