---
name: image-to-prompt
description: "Drop a reference image: palette, brightness, contrast, orientation and composition are read in the browser and written into a prompt skeleton for image-to-video. Use when someone asks for: image to prompt. Free browser tool on SuperDuperAI (superduperai.co): runs in the user's browser, nothing is uploaded."
---

# Image to prompt

Drop a reference image: palette, brightness, contrast, orientation and composition are read in the browser and written into a prompt skeleton for image-to-video.

Page: https://superduperai.co/tools/image-to-prompt/?utm_source=skill

## Limits

- Runs entirely in the user's browser. Nothing is uploaded, no account, no watermark.
- Because it runs on the user's machine, you cannot call it as an API: open the page for the person, or drive it with a browser (Playwright) on their machine.

## How to use it as an agent

1. The person has a file and a browser: give them https://superduperai.co/tools/image-to-prompt/?utm_source=skill and say what to set. The file stays on their device.
2. You run a browser (Playwright, computer use): open the page, put the file in the drop zone, wait for the result, save the download.

## SuperDuperAI MCP

Every tool on SuperDuperAI and our other sites, their skills and guides, as a read-only MCP server (`list_tools`, `get_tool`):

```sh
# Claude Code
claude mcp add --transport http superduperai https://superduperai.co/mcp
# Codex CLI
codex mcp add superduperai --url https://superduperai.co/mcp
```

Claude.ai, Claude Desktop, ChatGPT: add a custom connector with the URL `https://superduperai.co/mcp`.
Cursor and other clients (`mcp.json`): `{"mcpServers":{"superduperai":{"url":"https://superduperai.co/mcp"}}}`

## Related

- [Veo 3 prompt generator](https://superduperai.co/tool/veo3-prompt-generator/?utm_source=skill)
- [Frame grabber](https://superduperai.co/tools/frame-grabber/?utm_source=skill)

---
SuperDuperAI: https://superduperai.co/ · llms.txt: https://superduperai.co/llms.txt
