Photography AI - Professional Visual Engineering Skills Framework

Generate consistent AI images and cinematic sequences with seed and reference locking.

Updated Apr 12, 2026
One-click install
npx skills add https://github.com/marktantongco/brutalist-editorial --skill photography-ai-professional-visual-engineering-skills-framework
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: Photography AI - Professional Visual Engineering Skills Framework
Source: https://github.com/marktantongco/brutalist-editorial/tree/main/skills/photography-ai
Command: npx skills add https://github.com/marktantongco/brutalist-editorial --skill photography-ai-professional-visual-engineering-skills-framework

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill addresses the common failures of AI visual generation: inconsistent character identity across frames, photographic implausibility (lighting, lens, material response), and recurring artifacts like extra limbs, plastic skin, and temporal drift that break professional output quality.

Core Features & Use Cases

  • Structured Prompt Engineering: The Scaffold Method to compose subject, action, lighting, lens/specs, style, and quality for predictable renders.
  • Photographic Literacy: Guidance on lighting patterns, lens choice, aperture, and advanced rendering terms to produce physically believable images.
  • Strategic Negation & Identity Preservation: Negative prompting patterns and seed/reference locking to prevent artifacts and maintain character continuity across generations.
  • Post-Processing Workflow: Inpainting, upscaling, A/B iterative refinement, and external tool handoff for delivery-grade results.
  • Agent Orchestration: Design patterns for multi-agent production pipelines with human-in-the-loop checkpoints for scale projects.
  • Use Case: Produce consistent headshots for a brand campaign, orchestrate multi-scene cinematic sequences, or build production pipelines that hand off assets to color grading and upscaling tools.

Quick Start

Generate a photorealistic corporate headshot: 85mm f/2.8, Rembrandt lighting, seed 12345, preserve identity with a character reference and include a negative prompt to avoid plastic skin and extra fingers.

Frequently Asked Questions about Photography AI - Professional Visual Engineering Skills Framework

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I keep character identity consistent across multiple AI image generations?

To keep character identity consistent across AI image generations, use seed locking and reference image locking alongside negative prompting. This prevents temporal drift and maintains facial continuity across cinematic sequences or brand campaign headshots.

How do I write AI image prompts that produce physically believable lighting and lens effects?

To produce physically believable lighting and lens effects in AI image prompts, apply the Scaffold Method to structure subject, action, lighting patterns like Rembrandt, lens choice, aperture such as f/2.8, and style. This applies photographic literacy to your image generation process.

What is the best way to remove artifacts like plastic skin and extra fingers from AI photos?

The best way to remove artifacts like plastic skin and extra fingers from AI photos is applying strategic negative prompting. By explicitly defining unwanted elements in the negative prompt, you mitigate recurring generation flaws and improve output fidelity.

Can I build a multi-step AI image generation pipeline with post-processing checkpoints?

Yes, you can build a multi-step AI image generation pipeline using agent orchestration. This workflow supports human-in-the-loop checkpoints, inpainting, upscaling, and A/B iterative refinement to hand off delivery-grade assets to external post-processing tools.

Why do my AI generated headshots look unrealistic and how can I fix them?

AI generated headshots look unrealistic due to photographic implausibility in lighting, lens, and material response. Fix this by specifying camera specs like an 85mm lens, using explicit lighting patterns, and applying negative prompts to avoid plastic skin textures.