Plug-and-play template system · Built for Claude Code

One folder turns Claude into a cinematic production studio

One-time payment.  Instant access.  Use it on every campaign.

Drop the folder into Claude Code, follow the 12-step manual, and Claude becomes your prompt engineer, production manager, and quality controller — all at once. You give the direction. It runs the shoot, logs every result, and keeps going until the shot's right.

Runs on /Higgsfield CLI /Seedance 2.0 /Nano Banana /Claude Code

Prompting every shot by hand will break you

  • Your characters drift between shots. Faces, builds, and outfits change from one frame to the next.
  • Your hero product looks different in every render. Colour, shape, detail — none of it holds.
  • Reference images bleed into each other. Lighting and background leak across the whole scene.
  • You write, fire, and track every prompt by hand. Nothing's logged, so nothing gets better.

One good shot is a lucky generation. A campaign needs dozens of them — consistent, on-brand, every time. That's not a prompting problem. It's a production problem.

It's a whole production pipeline.
In one folder.

This doesn't treat your AI agent as a one-off image generator. It turns it into a proper cinematic automation engine that handles the whole job, start to finish. Claude takes three roles at once.

Role 01

Prompt Engineer

Writes, refines, and fires the prompts for stills and multi-shot video. You give the direction in plain English — it handles the prompt craft.

Role 02

Production Manager

Runs the Higgsfield CLI, captures your reference UUIDs, sorts the output folders, and keeps the shot tracker up to date.

Role 03

Quality Controller

Reads what came back, logs the failures, applies safe rewrites, and keeps going until the shot's approved.

Three moves from folder to footage

01

Drop it in

Add the folder to Claude Code. Drop your reference images into model ref/, product ref/, and environment ref/.

02

Follow the manual

Work through INSTRUCTION-MANUAL.md. Claude captures every Higgsfield UUID, writes ref-ids.md, and sets up the full pipeline.

03

Direct the shoot

Give direction like "Model 1 climbs the cliff at dusk, ending on a product reveal." Claude fires the prompts, reads the results, and iterates.

Claude drives the whole pipeline

The system connects to the Higgsfield MCP, fires the prompts, captures every reference UUID, and tracks each shot through to approval — without a single command typed by hand.

  • References captured. Every UUID logged to ref-ids.md as it runs.
  • Prompts fired through the CLI. Stills and video, optimised and logged.
  • Every shot tracked. Marked approved, needs iteration, or rejected in the shot tracker.
shot-tracker.csv Live
00:00 Shot 01 · EstablishingModel climbs the ridge at dusk Approved
00:06 Shot 02 · Push-inSlow move to the product reveal Approved
00:11 Shot 03 · DetailWind catches the fabric, logo holds Rendering
00:15 Shot 04 · WideDrone pull-back into sun flare Iterate

Real outputs from this system

Every frame below was made with this exact system — one campaign for a fragrance brand, built start to finish. Same model, same product, same world, holding consistent shot after shot.

Cinematic still — character portrait in the volcanic sulphur fields
Cinematic still — the hero fragrance bottle held over flowing lava
Cinematic still — product detail, sneakers on volcanic rock beside lava

Stills pulled from the 15-second multi-shot ads generated through the system.

Everything the studio runs on

Environment library

18 stunning environment images

Eighteen ready-made environment reference images come in the pack — fully realised worlds you can drop straight into a shoot or pull apart for your own. Each ships with a matching text descriptor, so you can set a scene without the background bleeding in.

Environment by text

No ref bleeding

Control lighting, mood, and atmosphere with descriptors instead of image refs. It sidesteps the common ref-bleeding problem completely.

Prompt kit

Prompt frameworks

Ready-made structures for stills and 5 to 15 second cinematic video with timestamped cuts, camera moves, physical sound design, lighting, and genre flags.

Pipeline

The production pipeline

Automatic ref-ids.md, prompt-log.md, seedance-failures.md, batched output folders, and a Google Sheets CSV shot tracker.

Guardrails

Content-safety guardrails

Pre-documented NSFW and IP triggers, safe rewrites, ref-stacking limits, and resolution rules so generations do not fail silently.

Feedback loop

It learns your brand

Claude reads previous results and iterates. After a few campaigns you hold a personalised record of what works for your brand.

Includes INSTRUCTION-MANUAL.md, HANDOFF.md, README.md, the prompt frameworks, the logging templates, the CSV shot tracker, and the model ref / product ref / environment ref / outputs folders. Twelve steps, start to finish. Swap the placeholders and references to spin up a new campaign in minutes.

Made for the work you actually do

Use 01

Fashion editorials

The same model and wardrobe held consistent across a full editorial set.

Use 02

Product launch films

A hero product that stays on-brand through every shot of the ad.

Use 03

Brand storytelling

Multi-shot narrative video with consistent characters and a controlled visual world.

Use 04

Solo creators & small teams

Campaign-grade output without hiring a full production crew.

Own the studio

Universal AI Cinematic Automation
The complete template system
$ 150

One-time payment · instant access · unlimited campaigns

  • The full 12-step INSTRUCTION-MANUAL.md
  • Still and 5 to 15s cinematic video prompt frameworks
  • Reference, logging, and failure-map templates
  • The Google Sheets CSV shot tracker
  • Content-safety guardrails and safe rewrites
  • HANDOFF.md, README.md, and all working subfolders
Send me the system

Instant access, all sales final

The moment you buy, the complete system is yours to download — every file, framework, and template. Because the full product is delivered on purchase, all sales are final and we do not offer refunds. Read the questions below, and if anything is unclear, reach out before you buy. We would rather answer your question than take a sale you are not sure about.

Before you buy

Do I need to know how to prompt? +
No. You give high-level direction in plain language. The agent writes, refines, and fires the optimised prompts for you. The frameworks are already built in.
What do I need to run this? +
An AI coding agent (the system is designed for Claude in Claude Code) and a Higgsfield account for the CLI. The system drives Seedance 2.0 for video and Nano Banana for stills.
Is this software I have to install? +
No. It is a structured template folder plus a 12-step manual. Nothing to install. You drop it into your agent, follow the steps, and start directing.
Can I use it for more than one campaign? +
Yes. It is modular and reusable. Swap the placeholders and reference images and spin up a fresh campaign in minutes. Every campaign also makes your personal what-works knowledge base stronger.
What is actually in the download? +
INSTRUCTION-MANUAL.md, HANDOFF.md, README.md, the still and video prompt frameworks, the ref-ids / prompt-log / failure-map templates, the Google Sheets CSV shot tracker, and the model ref, product ref, environment ref, and outputs subfolders.

Stop prompting shot by shot.
Start directing.

Hand the production work to your agent and keep the part that matters: the creative call.

Get the system — $150