Smart grocery comparison agent for Asian international students#1234
Merged
guihunwansui merged 8 commits intodevfrom Apr 3, 2026
Merged
Smart grocery comparison agent for Asian international students#1234guihunwansui merged 8 commits intodevfrom
guihunwansui merged 8 commits intodevfrom
Conversation
This feature enables DoWhiz to help users (primarily Asian international students)
compare grocery prices across multiple channels:
- Online: Weee, Yami
- Asian markets: H Mart, 168 Asian Mart
- Mainstream: Kroger, Costco, Sam's Club, Aldi
Key additions:
- grocery-comparison skill with detailed instructions for price comparison
- GroceryStore MongoDB module for persisting prices and user preferences
- User preference template for taste/distance/cultural factors
- Prompt.rs integration to enable the agent to use the skill
- Market research and competitor analysis document
The agent considers:
- Cultural taste preferences ("American cakes are too sweet")
- Distance and transportation constraints
- Store memberships (Costco, Sam's)
- Delivery vs in-store trade-offs
Co-Authored-By: Claude Opus 4.5 <[email protected]>
Added capabilities to mine price information from: - Google Maps reviews (price sentiment and specific mentions) - 小红书 (Xiaohongshu) - best source for Chinese grocery prices - WeChat articles and public accounts - Reddit and local forums - Yelp reviews Key additions: - Social media price discovery workflow - Search query templates for different scenarios - Data freshness guidelines and trust levels - Handling when no results are found - Tips for saving discoveries to user memory Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Remove unsupported allowed-tools frontmatter - Add 6-step onboarding questionnaire flow for collecting user preferences - Basic profile (ethnicity, location, household size) - Shopping habits (transportation, delivery preference, memberships) - Main shopping categories - Category-specific preferences (meat, snacks, vegetables, hotpot) - Taste profile (sweetness, spiciness, saltiness acceptance) - Budget and priority ranking - Add product evaluation extraction table for comprehensive reviews - 分量/规格, 品质评价, 加工难度, 口味描述, 性价比, 适合人群, 踩坑提醒 - Reference 小红书 "问一问" AI for quick product queries - Define memory storage format for questionnaire results Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Align template with new onboarding questionnaire structure - Add cultural background, household size, taste profile sections - Add category-specific preferences (meat, snacks, vegetables, hotpot) - Add budget mindset and priority ranking - Add product evaluation notes section for quality tracking - Include example notes from 小红书 reviews Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Create GroceryPreferencesQuestionnaire.jsx with 6-step wizard flow - Step 1: Basic profile (cultural background, location, household) - Step 2: Shopping habits (transport, preferences, memberships) - Step 3: Main categories selection - Step 4: Category-specific preferences (meat, snacks, vegetables, hotpot) - Step 5: Taste profile (sweetness, spiciness, saltiness sliders) - Step 6: Budget and priority ranking (draggable list) - Create accompanying CSS with modern styling - Add data source availability status to SKILL.md (tested 2026-04-02) - Weee: working - Google Maps: working - Kroger/Costco: blocked (need API/cookies) - Xiaohongshu: blocked (non-China IP) - Document UI component usage in SKILL.md Co-Authored-By: Claude Opus 4.5 <[email protected]>
Kroger API Integration: - Add kroger_api.rs with OAuth2 authentication - Support product search, store lookup, and product details - Automatic token refresh with expiry handling - Full type-safe response parsing Questionnaire Updates: - Add store selection UI with categorized groups: - Asian online (Weee, Yami) - Asian physical (168, H Mart, 99 Ranch, etc.) - Warehouse clubs (Sam's Club, Costco) - Mainstream US (Kroger, Walmart, Target, Aldi, etc.) - Require at least one store selection to proceed - Add Sam's Club to data source availability status Required env vars for Kroger API: - KROGER_CLIENT_ID - KROGER_CLIENT_SECRET Register at: https://developer.kroger.com/ Co-Authored-By: Claude Opus 4.5 <[email protected]>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Tested with credentials: - OAuth2 token: working - Product search: working - Store lookup: working (3 stores near Ann Arbor 48109) - Pricing: working when locationId specified Co-Authored-By: Claude Opus 4.5 <[email protected]>
- Extract product_data_to_kroger_product() helper in kroger_api.rs - Pre-compute store category lists at module level - Add useCallback to drag handlers in DraggablePriorityList Co-Authored-By: Claude Opus 4.5 <[email protected]>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Features
grocery_store.rs): Products, prices, stores, user preferenceskroger_api.rs): OAuth2 auth, product search, store lookupGroceryPreferencesQuestionnaire.jsx): 6-step onboarding wizardNew Environment Variables
KROGER_CLIENT_ID=xxx # From developer.kroger.com KROGER_CLIENT_SECRET=xxxTest Plan
cargo build -p scheduler_module🤖 Generated with Claude Code