leo414x/ai_cinematic_prompt — explained in plain English
Analysis updated 2026-05-18
Generate a cinematic quality prompt for a single AI generated image.
Transfer the visual look of a specific movie or show onto a new scene you describe.
Get a confirmable LOOK CARD before generating a full prompt for a vague style idea.
Build a gallery of consistent, film grade AI images using saved presets.
| leo414x/ai_cinematic_prompt | avbiswas/sam2-mlx | chanthruu/wind_energy_monitor | |
|---|---|---|---|
| Stars | 27 | 27 | 27 |
| Language | Python | Python | Python |
| Setup difficulty | easy | moderate | moderate |
| Complexity | 2/5 | 4/5 | 3/5 |
| Audience | designer | researcher | data |
Figures from each repo's GitHub metadata at analysis time.
Clone the repo into your agent's skills folder and restart the agent.
AI Cinematic Prompt is a skill built for AI coding agents like Claude Code and Codex. Instead of writing image generation prompts by hand, you install this skill and ask your agent to describe a scene, and it writes a movie quality prompt for you, styled after HBO shows, big budget television, and Hollywood films. One of its core features is look transfer: you can point to the visual style of a specific film, show, or cinematographer and apply that same look to a brand new scene you describe. The skill also uses presets that lock in camera choices, lens type, lighting setup, color grading, film grain, and mood, so the same style can be reused consistently across many prompts. If your request is vague about style, the skill will not guess. It first shows you a LOOK CARD, a short summary of the visual direction it plans to take, so you can confirm or adjust before it generates the full prompt. A lot of attention goes into avoiding what the README calls cheap AI image problems: plastic looking skin, an obvious CG feel, over sharpened detail, and lighting that has no real source. The skill applies graded anti slop rules to steer away from these issues, and it is tuned for popular image generators including GPT Image, Flux, and Midjourney, with attention to film grain, shadow noise, and realistic tonal depth. The newest version, 0.3.0, improves how fog, light beams, and a sense of depth are generated, choosing between contrast fading, atmospheric haze, or layered occlusion depending on the scene, rather than always applying generic fog. You install it by cloning the repository into your agent's skills folder, for example inside Claude Code or Codex. A demo gallery is included showing neon lit characters, large scale environments, dragon scenes, and space themed images. The license is restrictive: resale, renaming, and use inside paid commercial products or platforms are not allowed without written permission from the author. The project states it has no official connection to HBO, Warner Bros, A24, Netflix, or any camera, lens, or software brand it references for style purposes.
A Claude Code and Codex agent skill that writes Hollywood and HBO style cinematic image prompts, with look transfer between movies and reusable camera and lighting presets.
Mainly Python. The stack also includes Python.
You may use it personally, but you cannot resell it, rename it, or build it into a commercial API, agent platform, or paid product without written permission from the author.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly designer.
This repo across BitVibe Labs
double-check against the repo, no cap.