# Prompt enhancer

> Enhance rewrites a rough idea into an engine-grade video or image prompt — subject, motion, camera, light, mood in one paragraph — or invents one from scratch. Free, in the studio, the API and MCP.

Good prompts have a shape: subject, what moves, the camera, the light, the mood, in one flowing paragraph. Enhance takes whatever you typed and writes that paragraph in the register the engines like; Surprise me invents one from nothing. It is free, keeps your @mentions intact, and sits under every composer.

## What it changes

- Adds the missing layer: a draft with only a subject gets motion, camera and light.
- Removes the noise: adjective piles and camera jargon soup become one directed sentence.
- Adapts to the engine you picked — the phrasing that suits Seedance is not the phrasing that suits Kling.

## Where it lives

The sparkle button on the video, image and home composers; POST /v1/prompts/enhance in the API (free); the enhance_prompt tool on the MCP server, so ChatGPT or Claude can polish a prompt before rendering; `eroq enhance` in the CLI.

## Starting points

- **A draft to enhance** — woman on a rooftop at night, wind

## FAQ

### Is the prompt enhancer free?

Yes — in the studio, the API and MCP. Only the render that follows costs credits.

### Does it work for images too?

Yes: kind "image" writes a photographer's brief instead of a director's.

### Will it keep my characters?

@mentions are preserved verbatim so they keep binding to your cast and sets.

## Related

- https://eroq.ai/blog/ai-video-prompt-guide — Prompt guide
- https://eroq.ai/tools/text-to-video — Text to video
- https://eroq.ai/docs/enhance — API reference

---

This page as HTML: https://eroq.ai/tools/prompt-enhancer · Studio: https://eroq.ai/studio · Docs: https://eroq.ai/docs · Machine index: https://eroq.ai/llms.txt