From grocer
Use when the user needs to find products at local stores, build a hardware/shopping list, compare prices across retailers, or search for specific items. Triggers on shopping lists, product searches, "where can I buy", store availability checks.
How this skill is triggered — by the user, by Claude, or both
Slash command
/grocer:grocerThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Browse real retailer websites via Playwright MCP to build shopping lists with actual prices, stock status, and clickable links. Prefer local stores over online.
Browse real retailer websites via Playwright MCP to build shopping lists with actual prices, stock status, and clickable links. Prefer local stores over online.
digraph grocer {
"User shares items" -> "Create list file in ~/dev/grocer/";
"Create list file in ~/dev/grocer/" -> "Dispatch parallel agents per store";
"Dispatch parallel agents per store" -> "Item found?" [label="per item"];
"Item found?" -> "Record: name, price, URL, stock" [label="yes"];
"Item found?" -> "Try alternate keywords (3-5 variations)" [label="no"];
"Try alternate keywords (3-5 variations)" -> "Item found?" [label="retry"];
"Try alternate keywords (3-5 variations)" -> "Try other stores / Google fallback" [label="exhausted"];
"Try other stores / Google fallback" -> "Suggest DIY alternatives or online last resort";
"Suggest DIY alternatives or online last resort" -> "Update list";
"Record: name, price, URL, stock" -> "Update list";
"Update list" -> "Present organized by store";
}
~/dev/grocer/ folder — create if missing~/dev/grocer/config/preferences.md for user's location, favorite stores, language, and measurement preferences~/dev/grocer/config/stores/*.md for any store-specific tips the user has taught{project}-list.md with items, specs, and statusgrocer/sessions/{store}-{item}.mdMost retailer sites are JS-rendered — WebFetch returns empty HTML. Use Playwright MCP.
1. browser_navigate → store search URL
2. browser_snapshot → read search results
3. Extract: product name, price, URL, product code, stock status
4. browser_click → product page if needed for details
5. browser_snapshot → read product details
If Playwright fails or site blocks automation, fall back to Google: "PRODUCT" site:RETAILER.ca
https:// link — codes alone are useless)site: as last resortgrocer/{store}-{item}.md# Shopping List (DATE)
## Project: description
---
## BOUGHT
### Item Name — PURCHASED
- [**Product Name — price**](https://full-url)
- Specs, notes
---
## STILL NEED — STORE NAME (location)
### Item Name
- **Spec:** detailed requirements
- [**Product — price** ✓ PICK THIS](https://full-url)
- [Alternative — price](https://full-url) (why it's backup)
---
## SUMMARY
| # | Item | Where | Price | Status |
|---|------|-------|-------|--------|
| 1 | Item | [Store](url) | $X.XX | BOUGHT/NEED/CONFIRMED |
npx claudepluginhub maferland/grocer --plugin grocerResearches products across the web via Firecrawl search and scrape, comparing prices, specs, and reviews to produce a shopping recommendation or cart-ready summary.
Compares live prices across retailers (Amazon, Walmart, eBay, Best Buy, Google Shopping) to rank offers and recommend where to buy. Handles product names, ASINs, and URLs with region awareness.
Queries Bunnings NZ/AU product search, prices, SKUs, store details, category browse, and redemption specials via no-login CLI. Read-only. Useful for shopping assistance and product data retrieval.