The Agent Skills Directory

[SAFE]: The skill provides documentation and workflow instructions for interacting with Replicate's official API for running AI models via predictions, webhooks, and streaming.
[DATA_EXPOSURE]: The skill facilitates data transfer through HTTPS webhooks and API endpoints. This is a core functionality of the service and follows standard API integration practices.
[INDIRECT_PROMPT_INJECTION]: The skill processes outputs from model predictions which could contain instructions. However, the skill provides clear guidelines for validating input parameters against schema constraints (minimum, maximum, enum values) and emphasizes fetching the latest schemas to ensure validity.

run-models