Image-to-Text Contextmenu

About

Adds a menu item to the context menu for images. Clicking it analyses the image, and copies a textual description of the image to the clipboard.

If the image contains text, then that will be extracted too.

Examples:
- generating descriptions of photo's or artwork
- extracting data from screenshotted forms, spreadsheets, and charts and graphs
- listing labels, shapes, and relationships in diagrams

In the options page, you can create custom prompts to cover specific scenarios. Optionally, you can even specify a JSON schema to exert precise control on how items extracted from the image are represented in the output.

In addition to creating your own prompts, the options page also gives you access to an online prompt library. Use the online prompt library to quickly build a collection of custom prompts or to get inspiration for creating your own. 

WARNING: This plugin relies on experimental built-in AI features of google chrome. For now, they require you to enable a few feature flags. You can enable these by opening chrome and navigating to chrome://flags. There, look for "Gemini" and enable the following:

- "Prompt API for Gemini Nano"
- "Prompt API for Gemini Nano with Multimodal Input"

See https://developer.chrome.com/docs/ai/built-in for the latest news on these features.