jnMetaCode/ai-shortfilm-prompts

What it solves

This project provides a structured system for creating high-quality, cinematic AI video prompts. It moves away from vague descriptors like "cinematic" or "4K" and instead uses a technical, model-agnostic framework based on real-world cinematography (camera lenses, focal lengths, and physical imperfections) to achieve visceral realism in AI-generated short films.

How it works

The system utilizes a 5-stage prompt structure to ensure consistency and detail:

1. Core theme: Style tags separated by pipes (e.g., gritty hard sci-fi).
2. Character & scene: Specific details on appearance and environment.
3. Atmosphere & quality: Visual base, color tone, and style core.
4. Camera rules: Specific camera models, lenses (e.g., Panavision C-series), and movement (e.g., "breath-like float").
5. Storyboard: A per-second or per-shot breakdown of the action.

It is available as a prompt library, a browser-based prompt builder, and a Claude Code Skill (plugin) that automates the generation of these structured prompts while performing a 10-item self-check.

Who it’s for

AI filmmakers, content creators, and users of video generation models such as Sora, Kling, Veo, Seedance, and Runway, who want to move beyond generic AI aesthetics to professional film-grade output.

Highlights

• Cinematography-based prompting: Uses real camera and lens specifications to anchor AI models to actual movie aesthetics.
• Model-agnostic framework: Compatible with major 2026 video engines including Sora, Kling, Veo, and Seedance.
• Comprehensive library: Includes 21 genre templates and a wide array of worked examples (e.g., cyberpunk, claymation, FPV drone).
• Claude Code Integration: A dedicated plugin that turns ideas into model-ready prompts using the 5-stage method.