Recipe Leftover Suggestion Generator

WHAT IT DOES

This userscript helps you reduce food waste by suggesting delicious ways to repurpose leftover ingredients from any recipe. It scans the page for ingredient lists, then matches each ingredient against a built-in database of common leftover transformations. For example, if a recipe includes "roast chicken", the script suggests chicken salad, soup, tacos, pot pie, and sandwiches. If it finds "cooked rice", it recommends fried rice, rice pudding, stuffed peppers, or congee. The suggestions appear either as inline annotations directly below each ingredient or as a tidy list inside the panel, making it easy to plan meals around what you already have.

WHERE IT RUNS

The script works on every website you visit—food blogs, recipe aggregators, meal planning apps, cooking forums, or any page containing ingredient lists. It runs entirely within your browser, reading only the visible text and DOM structure of the current page. It does not track your browsing history, access other tabs, or send any data externally. All leftover transformation rules and ingredient databases are stored locally within the script—no external API calls are made.

HOW TO USE

After installation, open the panel by clicking the tool's icon in the BotGentz sidebar. Choose your preferred display style: "Inline suggestions" adds a small block below each matching ingredient with leftover ideas, while "List in panel" shows all suggestions in a scrollable list within the panel itself. You can also choose how many suggestions to show per ingredient—3, 5, or all available options. Click "Generate suggestions" to scan the page and apply the selected mode. If you want to remove all suggestions and return the page to its original state, click "Clear suggestions". Auto-scan on page load is enabled by default, so suggestions appear automatically when you visit a recipe page.

THE TECHNICAL MECHANISM

This script uses a multi-pass matching system to identify ingredients and generate relevant leftovers. First, it extracts ingredients by scanning for common structural patterns: bullet points, numbered steps, and container elements with class or ID names containing "ingredient" or "ingredients". Each candidate line is validated by checking for measurement units (cups, tbsp, oz, g, etc.) or natural-language patterns to avoid false positives. Second, the script normalizes each ingredient text by removing measurements, punctuation, and extra whitespace, then performs a two-stage lookup against the leftover database. The database contains over 60 entries organized by food type—each entry includes a list of keyword variants (e.g., "roast chicken", "rotisserie chicken", "cooked chicken") and an array of suggested dishes. The matching is case-insensitive and uses substring matching to catch variations like "chicken breast" matching "chicken". If an exact keyword match fails, the script falls back to partial word matching and singular root matching to maximize coverage. Once matches are found, the script either injects a styled container element next to the ingredient's DOM node or appends the suggestions to the panel's body, depending on the selected display mode.

THE PANEL

The panel is draggable anywhere on the screen and remembers its position between sessions. It can be collapsed to a small floating bubble to minimize distraction, and expanded again with a single click. The panel's width can be adjusted, and pressing the Escape key closes it instantly. All settings—your chosen display style, maximum suggestions per ingredient, and auto-scan preference—are persisted locally using GM_setValue, so your preferences remain intact across page reloads and browser restarts.

PLEASE NOTE

This userscript requires the free BotGentz extension to function. The script's leftover database covers approximately 16 major ingredient categories with over 100 suggestion combinations; however, not every ingredient will have suggestions—especially uncommon or specialized items. The ingredient detection works best on well-structured recipe pages with clear lists; dynamically loaded content or unusual formatting may reduce accuracy. No data from your browsing is stored or transmitted anywhere; all processing occurs entirely in your browser. The script never modifies any data except for adding visual suggestion elements next to ingredient containers.