The Agent Skills Directory

[SAFE]: The skill incorporates strong security and ethical guidelines, specifically prohibiting the generation of unpermitted likenesses or photoreal fake-event footage targeting real entities.
[SAFE]: A 'human-in-the-loop' requirement is enforced for judging all video generations and approving final exports before publishing, preventing autonomous misuse.
[EXTERNAL_DOWNLOADS]: The instructions reference the official lumaai-python SDK and link to lumalabs.ai for pricing and documentation; these are well-known, official services associated with the skill's primary purpose.
[COMMAND_EXECUTION]: The skill identifies that the agent can interact with the official Luma SDK (lumaai-python) where an API connection is established, which is appropriate for a tool-focused integration skill.

luma