Skip to main content
Glama

Build a shopping list

build_shopping_list

Compose a consolidated grocery list from 1-8 recipes in the collection and hand it off for shopping — nothing is stored; the list is an artifact of this conversation. ALWAYS call with mode 'propose' first and show the user the returned list. Only after they explicitly approve in a LATER turn may you call a send mode: 'cart' (adds the items to their own linked Kroger-family cart — Fred Meyer, Mariano's, Ralphs… — requires the grocery:cart permission; asks them to connect their Kroger account the first time), 'coupons' (Pete's Fresh COUPON mode — clips this week's matching specials to their loyalty card; it can ONLY match items currently on coupon, so most items staying unmatched is normal; same grocery:cart permission) or 'email' (mails them the list, forwardable to whoever shops — requires the email:send permission). NEVER propose and send in the same turn. Pass pantry staples they keep in 'exclude'. recipe_ids MUST be copied exactly from search_recipes or list_conceits results — never invented.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "recipe_ids": "<recipe_ids>",
      "mode": "<mode>"
    },
    "name": "build_shopping_list"
  }
}

Input Schema

NameRequiredDescriptionDefault
zipNocart mode: US zip for first-time store selection. Only needed when the tool asks for it — the store is picked from the user's location when that is known, and remembered after.
modeYespropose = show the consolidated list in chat (always first); cart = add items to the user's Kroger cart (only after the user approved the proposal in a previous turn); coupons = clip matching Pete's Fresh weekly specials to the user's loyalty card (same approval rule; coupon-only — most items won't match); email = send the proposal email (same approval rule); link = build the Walmart cart link now (same approval rule).
titleNoOptional list title in the user's words, e.g. "Sunday dinner with Ramona".
excludeNoIngredients to leave off — pantry staples the user keeps, or items they said to drop. Plain names, e.g. "soy sauce".
recipe_idsYes1-8 recipe point ids, exactly as returned by your tools.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / mode / description
      Previous value: -"propose = show the consolidated list in chat (always first); cart = add items to the user's Kroger cart (only after the user approved the proposal in a previous turn); coupons = clip matching Pete's Fresh weekly specials to the user's loyalty card (same approval rule; coupon-only — most items won't match); email = send the proposal email (same approval rule); link = create the Instacart page now (same approval rule)."New value: +"propose = show the consolidated list in chat (always first); cart = add items to the user's Kroger cart (only after the user approved the proposal in a previous turn); coupons = clip matching Pete's Fresh weekly specials to the user's loyalty card (same approval rule; coupon-only — most items won't match); email = send the proposal email (same approval rule); link = build the Walmart cart link now (same approval rule)."
  2. First observed

TDQS

A4.6/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Goes well beyond the annotations by disclosing per-mode side effects (cart writes to a linked Kroger-family cart, coupons clips loyalty-card specials, email sends mail), required permissions (grocery:cart, email:send), first-time account connection, and the known coupon-match limitation. The only mild ambiguity is 'nothing is stored', which describes the list artifact rather than the tool's side effects and sits in tension with readOnlyHint=false, though the mode descriptions make the mutating behavior clear.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Dense but every clause carries a rule — mode ordering, permission requirements, coupon caveat, id provenance. The critical constraint (propose first) is front-loaded. It reads as a long single block, which slightly hurts scannability, but little is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

No output schema exists, and the description still explains what the user sees (the returned list shown in chat), what each mode does, the permission prerequisites, and the first-time-store flow for cart mode. Nothing an agent needs to call this correctly is missing.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the schema already carries most parameter meaning (baseline 3). The description adds real constraints on top: exclude is for pantry staples the user keeps, and recipe_ids MUST be copied exactly from search_recipes or list_conceits and never invented — a correctness rule the schema does not express.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource (compose a consolidated grocery list from 1-8 recipes) plus the scope limit, and frames the output as a hand-off artifact. This clearly separates it from siblings like add_to_shopping_list, get_shopping_list, and update_shopping_list without opening their schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicit sequencing rules: ALWAYS call mode 'propose' first and show the list, only send after explicit approval in a LATER turn, and NEVER propose and send in the same turn. Each send mode is enumerated with its own condition and prerequisite permission, so the agent knows exactly when each path is valid.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources