The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external content which could theoretically harbor malicious instructions.
Ingestion points: User-provided prompts in the messages array and external images via image_urls in SKILL.md are processed by the EachLabs API.
Boundary markers: The instructional examples do not utilize specific delimiters to isolate user-provided content from agent instructions.
Capability inventory: The skill is limited to making network requests to the eachlabs.run and eachlabs.ai vendor domains. No local file system writes, administrative command execution, or other high-privilege operations are specified.
Sanitization: No input sanitization or validation logic is defined within the skill documentation.

photo-colorization