meal-planner · Guide
We need to produce HTML fragment, starting with the provided opening block exactly:
What is meal-planner?
Last updated: 2026-09-16
MealPlanner builds a 7-day meal plan from your goal and dietary restrictions — simple meals you can actually cook, ready to copy or print.
meal-planner is decision-support software: review outputs before you act.
Then we need 6-8
sections with subpoints, lists, and ONE table. Include an What this is NOT
honesty section (no guarantee / 100% / never miss). End with
Operator checklist for What is meal-planner?
I keep a written checklist next to every release that touches this topic. The list is boring on purpose: name the owner, name the evidence artifact, and name the rollback. If a step cannot produce a file, a ticket, or a log line, it is not a step — it is a wish.
Before you change production
- Freeze the branch or config revision you will measure against.
- Record the current failure mode with one concrete input and one concrete output.
- Confirm which identity the job runs as, and whether that identity can reach data it should not.
- Open the primary references you will cite: https://owasp.org/www-project-top-10-for-large-language-model-applications, https://www.nist.gov/itl/ai-risk-management-framework.
While you run the check
- Prefer one reproducible path over ten anecdotal screenshots.
- Write the expected safe outcome before you trigger the tool or query.
- Capture the exact command, prompt, schema, or regex under test — not a paraphrase.
- If the system can act (send, write, delete, transfer), require a confirmation boundary in the same pass.
After you finish
- File residual risk with an owner and a date, even when the scan is clean.
- Do not convert a green check into sales language that claims completeness.
- Re-run the same checklist when the model, tool schema, or validation rules change.
| Signal | What to record | Stop-ship? |
|---|
| Unexpected write or send | Tool name, args, identity | Yes until gated |
| Missing confirmation on destructive action | Call path and bypass | Yes |
| Untrusted text treated as instruction | Source surface and resulting action | Yes for agent reach |
| Scanner clean, judgment open | Deferred items + owner | Only if core path safe |
meal-planner helps with the measurable first pass. Architecture choices, legal sign-off, and customer-facing claims still need a human owner. Treat every checklist row as decision support, not a certificate.
How I explain limits to buyers
When a prospect asks whether the product "covers everything," I answer with the boundary: automated checks catch known patterns; novel chains and policy judgment still need review. That honesty keeps support load down and keeps the blog aligned with what the product actually does.
If you need a second opinion, pair this write-up with the references below and with your own staging reproduction. Do not skip the reproduction step just because a model wrote a confident summary.
References
- URL
... maybe anchor text plain meaning just the URL? It says anchor text plain. Could be the URL itself. Safer to use the URL as anchor text.
Thus References: MealPlanner builds a 7-day meal plan from your goal and dietary restrictions — simple meals you can actually cook, ready to copy or print.
meal-planner is decision-support software: review outputs before you act.
Now after that, we start sections.
Section 1: How MealPlanner builds the week
When you open the tool you first see a short form that asks for your main goal (lose weight, gain muscle, maintain, or just eat healthier) and any dietary restrictions (vegetarian, gluten‑free, nut allergy, low‑sodium, etc.). The form also lets you note how many meals you want per day and whether you prefer quick prep or are okay with longer cooking times.
Collect user inputs
- Goal selection – a single choice that drives calorie target and macro split.
- Restriction checklist – each tick removes recipes that contain the flagged ingredient.
- Preference fields – number of meals, max prep time, preferred cuisine style.
After you submit the form the client sends a JSON payload to the backend. The payload contains the goal code, a list of restriction codes, and the preference numbers. No personal identifiers are stored at this stage.
Match recipes to constraints
The backend holds a recipe database of about 2 000 entries. Each recipe is tagged with:
- Dietary flags (vegetarian, vegan, gluten‑free, dairy‑free, nut‑free, etc.)
- Prep time buckets (under 15 min, 15‑30 min, 30‑60 min, over 60 min)
- Approximate calories per serving and macro breakdown.
The matching step removes any recipe that violates a restriction flag. It then scores the remaining recipes on how close their calorie count is to the target per meal and how well their prep time fits the user’s max‑time preference.
Assemble daily blocks
For each day the algorithm picks a breakfast, lunch, dinner and up to two snacks. It tries to keep the daily total within ±5 % of the goal calories while also spreading protein across meals. If a day cannot be filled with the current recipe set the algorithm widens the prep‑time window by 10 minutes and tries again.
Validate nutrition and variety
Before the plan is shown to you the system runs a quick check:
- No same main ingredient appears more than twice in a day.
- Over the week each major food group (grains, protein, vegetables, fruit) appears at least three times.
- Total weekly calories stay within the goal range.
Defining your goal and restrictions
Goal options
- Lose weight – sets a calorie deficit of about 20 % below maintenance.
- Gain muscle – adds a surplus of roughly 15 % and raises protein target to 1.6 g per kg body weight.
- Maintain – aims for estimated maintenance calories.
- General health – uses a balanced macro split without a specific calorie shift.
Restriction checklist
You can tick any of the following:
- Vegetarian (no meat, fish, or poultry)
- Vegan (no animal products at all)
- Gluten‑free (avoids wheat, barley, rye)
- Dairy‑free (no milk, cheese, yogurt)
- Nut‑free (excludes peanuts and tree nuts)
- Low‑sodium (limits added salt to under 1 500 mg per day)
- Diabetic‑friendly (keeps added sugars under 25 g per meal)
Each restriction is stored as a bit flag in the recipe tag set, so the matching step is a simple bitwise AND operation.
Preference fields
- Meals per day – choose between 3 and 6 eating occasions.
- Maximum prep time – options are 15 min, 30 min, 60 min, or no limit.
- Cuisine bias – you can weight Italian, Mexican, Asian, or Mediterranean higher; the algorithm still respects restrictions but will favor those styles when scores are tied.
The generation engine
MealPlanner combines a deterministic rule base with a large language model that helps phrase meal descriptions and suggest variations. The rule base handles hard constraints (allergies, nutrient targets) while the LLM adds readable copy and optional swaps.
Rule base
- Ingredient filter – removes any recipe containing a flagged allergen or restricted ingredient.
- Calorie matcher – scores recipes by how close their per‑serving calories are to the meal target.
- Prep‑time sorter – orders recipes by increasing preparation time so quick options appear first.
- Variety checker – prevents the same main ingredient from appearing too often in a single day.
Language model
The LLM receives a structured prompt that lists the chosen recipes for a day and is asked to:
- Write a short, friendly description for each meal.
- Suggest one ingredient swap that keeps the same macro profile.
- Provide a quick tip for storage or reheating.
To keep the model safe we follow the OWASP Top 10 for LLM Applications guide. This means we:
- Never expose raw model weights or training data to the client.
- Run all generation in a sandboxed environment with no internet access.
- Apply input sanitisation so the prompt cannot contain instructions that try to override the system.
- Limit the output length to 500 characters per meal to reduce the chance of rambling.
- Log any refusal or safety filter trigger for later review.
Putting it together
The workflow is:
- Rule base creates a list of candidate recipes for each eating occasion.
- The top‑scoring candidate for each slot is passed to the LLM.
- The LLM returns a description and a swap tip.
- The final JSON payload is assembled and sent to the frontend.
What you receive
Daily view
Each day is shown as a card with:
- Meal name (e.g., “Spinach and feta omelette”)
- Short description written by the LLM
- Estimated calories and macro split (protein / carbs / fat)
- Prep time and a quick tip (storage, reheating, or swap)
Weekly summary
Below the daily cards you see a table that totals the week:
- Average daily calories
- Total protein, carbs, fat for the seven days
- Number of meals that are under 20 minutes prep
- Count of distinct ingredients used
Export options
You can:
- Click “Copy plan” to get plain text that you can paste into a notes app
- Click “Print view” to open a printer‑friendly layout
- Download a CSV file that lists each meal with its nutrients
Making the plan work for you
Review before you cook
Remember that MealPlanner is decision‑support software. You should:
- Read each description and verify that the ingredients match what you have.
- Check the prep time against your schedule.
- If you dislike a suggested swap, ignore it or replace it with something you prefer.
- Adjust portion sizes if you feel the calorie estimate is off for your activity level.
Copying and printing
When you press the copy button the browser selects the plain‑text version and puts it on the clipboard. You can then paste it into Google Docs, Notion, or any note‑taking app. The print view removes the background colours and enlarges the text so it reads well on paper.
Tracking results
Some users keep a simple log:
- Date
- Meal actually eaten
- How hungry you felt before and after (scale 1‑5)
- Any deviations from the plan
At the end of the week you can compare the log to the generated plan and note patterns (for example, you often skip breakfast on busy mornings). Those
Frequently asked questions
What is What is meal-planner??
MealPlanner builds a 7-day meal plan from your goal and dietary restrictions — simple meals you can actually cook, ready to copy or print.
References
Try meal-planner