An intelligent recommendation engine that transforms raw product data into curated shopping experiences based on intent and constraints.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install product-recommender
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install product-recommender using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Product Recommender?
The Product Recommender is a specialized skill designed for retail digital employees within the Openclaw Skills ecosystem. It moves beyond simple keyword searching by acting as a virtual personal shopper that understands nuanced human intent, such as gifting occasions, style preferences, and complex budget constraints.
By leveraging existing product knowledge bases and inventory data, this skill provides deterministic filtering and qualitative reasoning for its suggestions. It is engineered to handle the 'help me choose' phase of the customer journey, ensuring that every recommendation is backed by a logical 'why' that drives higher conversion rates and customer satisfaction.
Product Recommender Use Cases
- Gift finding for specific recipients and occasions (e.g., birthdays, anniversaries)
- Outfit pairing and complementary product suggestions (cross-selling)
- Budget-optimized shopping within strict price ceilings
- Comparative analysis between two or more specific products
- Handling price objections with value-based alternatives
How Product Recommender Works
- Intent Extraction: The skill parses user input to identify signals like budget, recipient, occasion, and specific constraints.
- Hard Filtering: Items are immediately eliminated if they exceed the maximum budget, lack required attributes (e.g., material), or are out of stock.
- Soft Scoring: Remaining items are ranked based on their suitability for the recipient, occasion tags, and style keyword overlap.
- Selection: The engine selects the top N products (defaulting to 3) for presentation.
- Response Generation: The skill formats the output with product names, prices, and a persuasive 'reason for recommendation' tailored to the user's intent.
Product Recommender Setup
To deploy this skill within your Openclaw Skills environment, ensure your knowledge base includes a populated products[] array.
# Install the product recommender skill dependencies
pip install -r skills/product-recommender/requirements.txt
# Ensure the filtering script is executable
chmod +x scripts/recommend.py
Product Recommender Data Schema & Taxonomy
The skill organizes recommendations using a structured scoring system and maps to the following metadata taxonomy:
| Attribute | Description |
|---|---|
price / sale_price |
Used for hard budget filtering |
suitable_for |
Demographic and recipient mapping |
tags |
Occasion and style keyword matching |
stock_qty |
Availability check (if inventory integration is active) |
sales_rank |
Popularity weighting for soft scoring |
Product Recommender Advanced Features
- Upsell Logic: Automatically suggests a premium option (up to 20% over budget) if it offers a significant quality upgrade.
- Comparison Engine: Generates dynamic comparison tables when users are torn between specific items.
- Fallback Management: If no perfect match exists, it identifies the 'closest available' option to prevent dead-end conversations.
- Multi-lingual Intent Support: Native support for both English and Chinese shopping terminology.
SKILL.md
Loading
Related Openclaw Skills
TokenTracker
A comprehensive monitoring tool for tracking AI model token consumption and costs with automated reporting and threshold alerts.
58tongcheng
A specialized automation skill for extracting and organizing public merchant data, service rules, and promotional activities from the 58.com local services platform.
SVG Cover Generator
A professional vector artwork engine that generates polished, text-editable SVG covers for reports, social media, and landing pages.
IPinfo.io
A comprehensive IP intelligence skill that provides real-time geolocation, ASN, and organizational data for any IP address.
Feed Ads (In-Feed Advertising SOP)
An actionable standard operating procedure for executing, monitoring, and optimizing in-feed advertising campaigns with high efficiency.
Get My IP
A specialized skill designed to instantly fetch the machine's current public IP address using the ifconfig.me service.
Featured*
