# Reference Breakdown — prompt skill for prompts by model

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

Turn a frame you have rights to into a shot spec — size, angle, lens, light, palette — plus prompt variants.

- **Type:** [Prompt](https://skills.varg.ai/prompts.md) · [Prompts by model](https://skills.varg.ai/browse/skills/prompting.md)
- **Slug:** `craft-reference-breakdown`
- **Status:** draft
- **Publisher:** varg (varg, verified)
- **License:** MIT
- **Source:** https://github.com/vargHQ/skills
- **Works in:** Claude Code, Claude.ai, Codex, Cursor, Gemini CLI, OpenClaw, Any MCP client, varg Studio
- **Cost:** Free to analyze; any generation it proposes is billed in varg credits after estimate_cost.

## What it does

The popular 'still → reverse-engineer the lens, light, color temperature and composition → prompt variants' loop, done safely: the user supplies a frame they have rights to, and the skill describes it in technique-card ids and writes a shot spec plus variants per model. The image is analyzed in the session only — never stored in the skill and never sent to a generation provider by the skill — and the output is words and card ids, not a copy of the frame. Draft until the rights-attestation step and the evals are built.

## Use when

- User uploads their own photo or frame and asks how to get that look
- Matching a new shot to a campaign frame the user owns

## Not for

- Film stills the user does not own as image-to-video inputs
- Recreating a specific film scene, character or actor

## Install

Not published as a standalone skill yet. Source: https://github.com/vargHQ/skills

**Models it drives:** seedance_2_5, kling_v3, veo_3_1, omni_flash_1_1, nano_banana_2_1

## Example output

https://skills.varg.ai/previews/craft-reference-breakdown.webp — The kind of frame Reference Breakdown starts from, one you have rights to: an eye-level close-up on a long lens, warm backlight rimming the hair, a soft frontal key, a bokeh background and an out-of-focus foreground object in an amber-brown palette. The skill's own output is a shot spec in words and card ids, not this image. Tears of Steel (2012) frame at 09:15.4, 720p release from the Blender mirror (download.blender.org/demo/movies/ToS), full 2.40:1 frame. (credit: Tears of Steel (CC BY 3.0) — (CC) Blender Foundation | mango.blender.org, https://mango.blender.org/)

## Used for

- [Remake a reference video](https://skills.varg.ai/use-cases/remake-reference.md): Take a reel or ad you like, break it down shot by shot, and remake the format for your brand.

## More prompts by model skills

- [fal Prompting](https://skills.varg.ai/skills/fal-prompting.md): Prompt specific model families well: Kling, GPT Image 2 and Happy Horse guides, loaded only when needed.
- [Replicate Prompt Images](https://skills.varg.ai/skills/replicate-prompt-images.md): Write better image prompts and edits: photographic language, lighting, text rendering, consistent characters.
- [Replicate Prompt Videos](https://skills.varg.ai/skills/replicate-prompt-videos.md): Prompt video models like a director: shots, camera motion, four audio layers, dialogue, time-coded beats.
- [Video Prompting Guide](https://skills.varg.ai/skills/inference-sh-video-prompting-guide.md): Write prompts that video models follow: shot types, camera moves and per-model tips from Veo to Sora.
- [FLUX 3 Prompt Doctor](https://skills.varg.ai/skills/flux-3-prompt-doctor.md): Diagnose a FLUX 3 video brief before spending: resolve missing, conflicting or schema-changing asks.
- [FLUX Image Best Practices](https://skills.varg.ai/skills/flux-image-best-practices.md): Prompt FLUX image models well: structure, JSON prompts, hex colors, typography and multi-reference edits.