The Agent Skills Directory

[SAFE]: The skill interacts with eachsense-agent.core.eachlabs.run and storage.eachlabs.ai, which are verified infrastructure domains belonging to the skill's author, eachlabs.
[SAFE]: Authentication is handled via the $EACHLABS_API_KEY environment variable in the documentation examples, which is the recommended practice for secret management and prevents credential exposure.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data via image_urls and user-supplied messages to generate digital twins.
Ingestion points: External images are fetched via URLs provided in the image_urls field of the API request (SKILL.md).
Boundary markers: Data is isolated within a structured JSON payload, providing clear separation between instructions and data.
Capability inventory: The skill uses network operations to communicate with the EachLabs API for image and video generation.
Sanitization: The skill includes explicit guidance in the 'Error Handling' and 'Privacy & Consent' sections regarding content policy violations and the requirement for authorized use, which serves as a mitigation layer.

digital-twin-generation