The Agent Skills Directory

[SAFE]: The skill implements secure credential management by requiring the FAL_KEY or FAL_API_KEY to be provided through environment variables. No hardcoded secrets were found in the provided files.
[SAFE]: Network communications are restricted to legitimate and well-known service domains (queue.fal.run and api.fal.ai) for the purpose of submitting inference jobs and retrieving generated media assets.
[SAFE]: The scripts use standard Python libraries for file handling and networking. Media downloads are performed into user-specified output directories as part of the primary intended functionality of the skill.
[SAFE]: The prompt processing logic is consistent with the skill's purpose of interfacing with generative models. While it ingests external text, the data is treated as opaque payloads for the remote API and does not influence local command execution flow.

fal-video