meal-planner
Skill8 starsChanged 8 months agoMITView on GitHub
Plan meals, search Frisco for matching products, build a cost-effective cart with nutrition tracking, and iterate with the user to optimize.
8 other projects use a version of this, 1 of them edited.
Adopted by
Most recently taken up by
- sudeepban/claude-skillstheir own version, first seen Oct 20, 2026
- marco308/mealstheir own version, first seen Oct 20, 2026
- GarphenGate/moltline-osstheir own version, first seen Sep 19, 2026
- BryceYuuu/open-workflow-skillstheir own version, first seen Sep 5, 2026
- NinaKuskova/time-saver-aitheir own version, first seen Aug 19, 2026
- Itsme23476/everyday-claude-skillstheir own version, first seen Aug 5, 2026
Other versions8 projects use a version of this.
Word for word · 0
None found.
Edited · 1
- wydrox/martmart-cli★ 169% same
Wrote their own version · 7
- marco308/meals★ 5same name
- BryceYuuu/open-workflow-skills★ 1same name
- Itsme23476/everyday-claude-skills★ 1same name
- zwjbendn/weight-loss-skill★ 1same name
- GarphenGate/moltline-oss★ 0same name
- NinaKuskova/time-saver-ai★ 0same name
- sudeepban/claude-skills★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
The file
We split this skill into the rules it gives an agent
12 rules. 0 are common, 9 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 7 edited versions we read keep these.
9
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 13
If binary missing: make build.
- Very specific: line 28
Always use --category-id to narrow results.
- Distinctive: line 44
Use subagents in parallel to search multiple product groups at once (e.g., meat, dairy, vegetables, pantry).
- Distinctive: line 46
When a product is unavailable, search for alternatives and propose a substitution — update the meal plan accordingly.
- Very specific: line 53
Wait for explicit confirmation: before running without --dry-run.
- Distinctive: line 65
Add a macro summary table to each day in the plan.
- Distinctive: line 66
Add a weekly average at the bottom.
- Distinctive: line 81
After the cart is filled, review it with the user:
- Distinctive: line 105
After every change, show the updated cart total.
- Distinctive: line 117
Substitutions: update recipe text when swapping products.
- Distinctive: line 125
After the user places an order, review what was actually ordered:
- Distinctive: line 138
Use this to verify the order matches the plan and to inform future planning sessions.
“Common” is measured: kept by at least 6 in 10 of the 7 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Frisco Meal Planner
Help the user go from meal ideas or dietary guidelines to a filled Frisco.pl cart with full nutritional breakdown.
Prerequisites
./bin/frisco session verify— if it fails, guide the user throughsession loginorsession from-curl.- If binary missing:
make build. - Gather context from the user:
- Dietary guidelines, restrictions, or just meal ideas.
- How many people and which days (e.g., "weekend for 2").
- Budget expectations (if any).
- What they already have at home (oils, spices, staples) — ask early to avoid buying things they don't need.
Phase 1: Meal Plan
- Build a plan matching the user's input (number of days, meals per day, portions).
- Save to a markdown file (not in the repo — ask user where they want it, e.g., Obsidian vault).
- The plan is a living document — update it whenever the cart changes.
Phase 2: Product Search
Always use --category-id to narrow results. Without it, searches return irrelevant matches. Full category tree is in categories.md at the project root.
# Best-match pick (scores by name match + price/kg + pack size):
./bin/frisco products pick --search "PHRASE" --category-id CATID --top 3
# Raw search when pick isn't enough:
./bin/frisco products search --search "PHRASE" --category-id CATID --format json 2>/dev/null | \
jq '[.products[] | select(.product.isAvailable == true) |
{id: .productId, name: .product.name.pl, price: .product.price.price,
grammage: .product.grammage, unit: .product.unitOfMeasure}] | .[0:5]'
# Add single product by search:
./bin/frisco cart add --search "PHRASE" --category-id CATID --quantity NUse subagents in parallel to search multiple product groups at once (e.g., meat, dairy, vegetables, pantry). Each subagent uses the Go CLI + jq.
When a product is unavailable, search for alternatives and propose a substitution — update the meal plan accordingly.
Phase 3: Cart Assembly
- Collect all product IDs and quantities into a JSON file.
- Dry-run first:
./bin/frisco cart add-batch --file FILE --dry-run - Show the user a full summary table with names, prices, and quantities.
- Wait for explicit confirmation before running without
--dry-run.
Phase 4: Nutrition / Macros
The user cares about nutritional values. For every meal plan, provide per-meal and per-day macro breakdowns (kcal, protein, fat, carbs).
Data sources (in priority order):
- Frisco API —
./bin/frisco products nutrition --product-id PID --format jsonreturns per-100g values for many products. Use subagents to batch-query nutrition for all key ingredients. - Standard nutritional databases — for products where the API returns no data, use well-known reference values. Mark these clearly as "estimated" in the plan.
How to present:
- Add a macro summary table to each day in the plan.
- Add a weekly average at the bottom.
- Note which values come from the API vs estimates.
- Flag any days that seem too high or too low for the user's goals.
# Query nutrition for a product:
./bin/frisco products nutrition --product-id PID --format json 2>/dev/null | \
jq '{kcal: (.nutrition.nutriments[] | select(.name.pl | test("kcal")) | .in100Gramms),
protein: (.nutrition.nutriments[] | select(.name.pl | test("Białko")) | .in100Gramms),
fat: (.nutrition.nutriments[] | select(.name.pl | test("^Tłuszcz")) | .in100Gramms),
carbs: (.nutrition.nutriments[] | select(.name.pl | test("^Węglowodany")) | .in100Gramms)}'Phase 5: Cost Optimization
After the cart is filled, review it with the user:
# Most expensive items:
./bin/frisco cart show --sort-by total --top 20
# Or via JSON for detailed analysis:
./bin/frisco cart show --format json 2>/dev/null | \
jq '[.products[] | select(.quantity > 0) |
{name: (.product.name.pl // .product.name.en), price: .product.price.price,
qty: .quantity, total: (.product.price.price * .quantity), grammage: .product.grammage}] |
sort_by(-.total) | .[0:20]'Propose savings by working through these strategies:
- Things the user already has — pantry staples accumulate. Ask.
- Frozen vs fresh — seasonal pricing matters. Frozen fruits/vegetables can be much cheaper.
- Waste reduction — if a package is much larger than needed, either spread it across multiple days in the plan or find a smaller alternative. Fewer product types with full utilization beats variety that rots.
- Duplicates — parallel searches can pick different products for the same ingredient.
- Quantity right-sizing — don't order 2 packs when 1 covers the week.
- Cheaper alternatives — different brands, non-BIO, different cuts, store brand.
- Better bought elsewhere — some niche items are overpriced on Frisco vs pharmacies or specialty stores. Flag these.
After every change, show the updated cart total.
./bin/frisco cart remove --product-id PID
./bin/frisco cart add --product-id PID --quantity N
./bin/frisco cart remove-batch --product-ids PID1,PID2,PID3Phase 6: Plan Reconciliation
The meal plan must always reflect what's actually in the cart:
- Substitutions: update recipe text when swapping products.
- Removed items: rework affected meals to use remaining ingredients.
- Portion changes: recalculate when some days serve more people.
- Ingredient budget: for items with large packages, add a table showing which days use what amount, so nothing goes to waste.
- Macros: recalculate nutrition whenever ingredients change.
Phase 7: Order Review
After the user places an order, review what was actually ordered:
# List recent orders:
./bin/frisco account orders list
# See products in a specific order:
./bin/frisco account orders products --order-id ORDER_ID --sort-by total
# Full order details:
./bin/frisco account orders get --order-id ORDER_IDUse this to verify the order matches the plan and to inform future planning sessions.
Principles
- Use the Go CLI —
products pick,products search,cart add --search,cart show --sort-by. No external scripts. - Confirm before cart changes — always show what will happen and wait for OK.
- Minimize waste — adapt the plan to use what's bought in full.
- Respect dietary restrictions — never substitute excluded ingredients, even if cheaper.
- Track nutrition — the user wants to know what they're eating. Include macros in every plan.
- Iterate — present, discuss, adjust. Don't aim for perfection in one pass.
- Show the running total — the user should always know where they stand.
- Keep personal files out of the repo — plans, shopping lists, cart JSONs go where the user wants them (e.g., Obsidian vault), not in the git repo.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

