Generate 30 diabetes-friendly recipes from your own ingredient list with Claude
- Posted
- by Edwin van Wijk (DefaultConstructor)
- Last checked
- Method:
- How we verify
Evidence: Described by posterThe person describes what happened in words, without screenshots or video.
Context: Over the same period, the poster also started exercising more and made drastic diet changes.
Outcome
After a type 2 diabetes diagnosis, Edwin van Wijk had the Claude CLI build a categorized list of ingredients available in a standard Dutch supermarket, removed everything disliked (cheese included), then wrote goals and constraints in a CLAUDE.md file. Claude generated 30 varied recipes plus a table showing each one's main protein, carbohydrates and why it suits a diabetic. Each recipe van Wijk cooks gets printed for the recipe binder in the kitchen. Alongside exercising more and other diet changes, van Wijk reports losing almost 4 kg in four weeks and a glucose drop from around 15 to around 6 mmol/L.
AI tool
Prompt Recreated prompt
The source didn't share its exact prompt, so we wrote this from what it describes.
Using only the ingredients in ingredients.md, generate recipes that are healthy for a person with type 2 diabetes and taste good. Add a summary table with each recipe's main protein, carbohydrates and why it's suitable for a diabetic.
Try this in:
Sends the Recreated prompt shown above. ChatGPT and Grok open with it filled in, and ChatGPT may send it straight away. Gemini opens empty, so we copy the prompt for you to paste. Every button also copies it.
The source used Claude. These are general chat tools, so results may differ.
Opens in a new tab. We don't see what you send.
These links are only a shortcut. We're not affiliated with OpenAI, Google or xAI, and your results may not match the source.
Optional stronger version
Our rewrite with extra checks. The poster didn't test it.
Before you start, ask me 1–3 short questions about the details below and wait for my answers. Once I've answered, confirm the plan in one line and run the whole task start to finish. Goal: Create a set of recipes that fit my diet plan and only use ingredients I like and can buy locally. Questions to ask me first: - What diet guidance has my doctor or dietitian given me (carbohydrate targets, foods to limit)? - What equipment do I have, and how much time do I have to cook on weeknights? - How many people am I cooking for? Inputs (fill in or confirm): - My doctor's or dietitian's guidance: [DIET GUIDANCE] - Ingredients I can buy and like: [INGREDIENT LIST] - Ingredients to avoid: [DISLIKES AND ALLERGIES] - Kitchen equipment and time: [EQUIPMENT / TIME] - Servings: [SERVINGS] Rules: - Use only ingredients on my list. - Follow my doctor's or dietitian's guidance; don't set medical targets yourself. - Give each recipe exact quantities per serving so it scales easily. - Estimate carbohydrates per serving and label them as estimates. - Don't give advice about diabetes medicines; that's for my doctor. Checkpoint: Show me 5 recipes and the summary table format first, and wait for my feedback before writing the rest. If you get stuck: If my list is too short for variety, suggest a few ingredients to add and wait for my OK. Verify before you finish: Check every recipe against my ingredient list and dislikes, and recheck the carbohydrate estimates. Output: [NUMBER] recipes in a printable format, plus a summary table with main protein, estimated carbohydrates per serving and prep time. Done when: I have a varied set of recipes I'll actually cook, all within my plan.
Source
https://www.defaultconstructor.com/2026/07/ai-as-my-personal-chef/
Related cases
- Understand flagged blood test results while waiting for a doctor with ClaudeRecreated promptEvidence: Article or interviewBeginner-friendly
When Judith Miller, a 76-year-old Milwaukee resident, saw routine blood work posted online, Miller messaged the doctor to ask about an elevated carbon dioxide reading and a 'low anion gap'. While waiting for a reply, Miller put the results into Claude and asked it to evaluate them. 'Claude helped give me a clear understanding of the abnormalities,' Miller told NPR, and because it didn't report anything alarming, the wait for the doctor's answer wasn't an anxious one. NPR's article also covers what to keep in mind before sharing lab results with a chatbot.
Claude · Personal · Source date:
- Research a Victorian-born grandfather's life with Claude, ChatGPT and Gemini deep research (partial)Recreated promptEvidence: Article or interviewPartial
Dan Taylor-Watt gave Claude, ChatGPT and Gemini the same brief, with deep research turned on, to research a paternal grandfather, Cecil, who died in 1960, starting from a Word document written by Taylor-Watt's father. Claude took about 6 minutes, used 340 sources and wrote an 11-page, 4,219-word report; it found no significant new information but flagged inaccurate dates, and it later made a 3-minute-30-second narrated video using code and an open-source voice model. ChatGPT took 9 minutes, ran 714 searches and wrote 14 pages (4,659 words); it caught an error about the year of the fire that destroyed service records, noted that the LNER railway wasn't created until 1923, found information on the first wife's family, and spotted a Lance-Corporal's chevron in a photo, but its video attempt fell short. Gemini wrote the longest and most visual report (5,347 words), but its maps were wrong (Osmotherley is east, not south, of Northallerton), it missed the date problems, and it said it couldn't make a video.
Claude, ChatGPT, Gemini · Personal · Source date:
- Audit a Proxmox home lab with Claude and find skipped backups, stale IPv6 settings and a stuck updateRecreated promptEvidence: Article or interview
Adam Davidson gave Claude access to two Proxmox machines in a home lab and asked it to look for problems without changing anything. Writing for How-To Geek, Davidson says Claude found three overlooked problems. The backup job only covered services Davidson had picked by hand, so newly added services were being skipped; the fix was to back up everything by default and exclude what isn't needed. The ISP had changed the IPv6 prefix, so a few services still pointed at a Pi-hole IPv6 address that no longer existed, which was why a Jellyfin list plugin had been silently failing. And on the OptiPlex running Proxmox, the enterprise repository was off with no free repository turned on, so it was stuck on 9.1.1; it has since been updated.
Claude · Personal · Source date: