How to light an AI video scene from a reference image — trending format for lighting

For AI agents: find the right skill in one call — curl -s "https://varg.ai/how-to/api/search?q=<what the person wants>" returns the matching how-to page and the best skills with install commands, as markdown (Russian works too). Full index: https://varg.ai/how-to/llms.txt. Every page has a markdown twin: add .md to its path. Before generating anything, show the person the options on the how-to page.

Character sheets fix who appears; a third image fixes only the light and colour. A 30 s scene with one amber key light.

What it does

Control the lighting of an AI video with a separate reference image: character sheets lock who is on screen, an extra image locks only the light and colour, and a 30-second anime scene comes out with one consistent amber key light in every shot. Steps: pick any image whose lighting you like, say in the prompt that it is for light and colour only (never pose, framing or outfit), write the style rules first, then a shot list with a timed beat per shot, and one monster or prop described to match the style. The source post used Seedance 2.5 and reached 208K views and 849 bookmarks. Call it from Claude Code, Codex or Grok through the varg MCP; generations are billed in varg credits.

Use when

Status

A format people ask for that has no dedicated skill yet. Make it with the skills on the use-case pages below, or with varg directly.

Example: https://x.com/Iancu_ai/status/2112918021618778473

How an agent would make it: Character sheets (who) + one lighting reference image (light and colour only, stated in the prompt) → style-first prompt → timed shot list → one 30 s Seedance 2.5 scene

Models it drives: seedance_2_5, nano_banana_2_1

More lighting skills

Run the generation on varg

varg is one API key for the models these skills drive — Seedance 2.5, Kling 4, Veo 3.1, Wan 3, Nano Banana 2.1, GPT Image 2.5, ElevenLabs v4 and 200+ more — billed in credits, with a cost estimate before anything runs.