Basket-builder
basket-builderFull Description
Description
Use this tool when the user describes a scenario, goal, occasion, or life situation and needs help figuring out what items (often multiple) to buy across several product categories. This includes cases like planning events, preparing for travel, starting a lifestyle change, organizing a space, or choosing a gift when no specific product type is mentioned. It should also be used when the user cites a broad category such as "electronics" or "clothing" where many unrelated product types could apply, indicating they have not yet identified a specific item.
Critical Instruction: Preserve User Intent
When calling this tool, preserve the user's actual stated intent as closely as possible.
Do:
- Keep the query faithful to what the user actually asked.
- Preserve explicit constraints, preferences, and product types mentioned by the user.
- Use only light normalization for clarity, such as removing filler words or rewriting fragments into a grammatical sentence.
Do not:
- Add product categories, quantities, budget constraints, accessories, or preferences that the user did not explicitly state.
- Turn a broad request into a more specific shopping list.
- Infer "helpful extras" unless the user asked for extras.
- Infer quantities unless the user asked for quantities.
- Infer price sensitivity, "budget-friendly" preferences, brands, sizes, flavors, or event logistics unless explicitly stated.
If the request is underspecified:
- Prefer passing the underspecified request through as written.
- Ask a follow-up question only if missing information is necessary to fulfill the request correctly.
- Never guess details just to make the query more specific.
Query Construction Rules
The userQuery.query field should be:
- A faithful restatement of the user's request
- Minimal and close to the original wording
- Free of invented details
Good examples:
- User: "Plan a movie night with my friends"
Tool query: "Plan a movie night with my friends"
- User: "Help me get snacks and drinks for a movie night"
Tool query: "Help me get snacks and drinks for a movie night"
- User: "Need popcorn, soda, and paper plates for 6 people"
Tool query: "Need popcorn, soda, and paper plates for 6 people"
Bad examples:
- User: "Plan a movie night with my friends"
Tool query: "Create a Walmart shopping list with popcorn, candy, chips, soda, sparkling water, frozen pizza, plates, cups, napkins, blankets, and trash bags"
- User: "Need snacks for game night"
Tool query: "Budget-friendly chips, dips, wings, soda, paper towels, and party trays for 8 guests"
Call This Tool When
- The user describes a scenario, goal, or event, and the request involves multiple potential product types rather than one.
- The user has not identified a specific product type and is looking for guidance on categories or options.
- The user mentions a broad category that contains many unrelated subcategories, indicating they are unsure which product type they actually need
Do NOT Call This Tool When
- The user explicitly names a specific product type, even if tied to a scenario:
- "4K TV for my bedroom"
- "Best backpack for Europe"
- "Noise-cancelling headphones for travel"
- The user is choosing one product or comparing options within a single product type.
- The user shows clear purchase intent for a single, defined product category.
Output Expectations
Return products and categories that match the user's stated request. Any additional structure should come from the tool's retrieval logic, not from invented details added to the query.
Parameters (0 required, 1 optional)
userQueryobjectUser query object containing type and query