You are helping me build the following AI agent workflow.

## Goal
Meal Planner & Cost Tracker: Submit a dish name to log ingredient costs in Sheets, get leftover recipes, receive nutritional advice, and draft an X post.

## Specification
- What it does: Streamline meal planning by automatically tracking ingredient costs, suggesting recipes for leftovers, and logging nutritional data. It analyzes your input dish to extract details, minimizes food waste with new ideas, and creates engaging social media posts ready for sharing.
- Trigger: Triggered by an event (Manual · on new menu item)
- Autonomy: Fully hands-off
- Expected setup effort: about an afternoon
- Tools/services involved:
  - Google Sheets: Storing recipes, costs, leftovers, and nutrition data
  - OpenRouter Chat Model: Analyzing ingredients and generating content
  - X (Formerly Twitter): Sending social media posts via direct message

## Known pitfalls, handle each one explicitly in your implementation
1. Ingredient pricing varies by region so include a disclaimer or allow manual price overrides to keep estimates realistic.
2. Image recognition accuracy may vary with complex dishes, add a fallback text input field for users to correct ingredient lists.

## Reference implementation
https://n8n.io/workflows/10967 (workflow template)
Fetch and inspect this before building. If it matches my stack, adapt it;
if not, rebuild the pattern with my tools.

## Process requirements
1. Before building: ask me which of the listed tools I actually use and
   what my platform is (n8n / Make / code / other). Do not assume.
2. Adapt the pattern to my answers; do not force the reference stack.
3. Address every pitfall above; tell me how you handled each.
4. Provide a test plan I can run before letting this touch real data.
5. Ask before any step that sends messages, modifies data, or spends money.

Source: https://usecasesforagents.com/use-case/meal-planner-cost-tracker/ via usecasesforagents.com