The Agent Skills Directory

[SAFE]: The skill implements secure credential management by instructing the agent to read the NOVOADS_API_KEY from a local .env file and providing explicit warnings against printing or committing the key to version control.
[SAFE]: Network operations are performed via curl targeting the vendor's official API at api.novoads.ai. The ability to override the host via NOVOADS_BASE_URL is a standard developer feature for API integrations and does not constitute a security risk in this context.
[INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for processing user-supplied images and text descriptions into motion graphic parameters.
Ingestion points: User-provided images and animation instructions within the conversation context.
Boundary markers: The skill uses a structured prompt template for the AI model but does not define explicit delimiters or sanitization for untrusted input.
Capability inventory: Uses curl for API interactions (estimates, uploads, and video generation).
Sanitization: No explicit sanitization of user-provided strings is mentioned before they are quoted in the API payload.
[SAFE]: Instructions to override specific API-level linter warnings (such as missing_actor_descriptor or banned_polish) are provided as functional guidance for the motion graphics domain to ensure design fidelity and do not target the agent's core safety protocols.

image-to-motion