An open-source, local-first market basket analysis tool designed to extract product association rules and business strategy recommendations from Excel data using the Apriori algorithm.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pairpulse
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 pairpulse 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 pairpulse?
pairpulse is a high-performance, developer-friendly command-line tool designed to simplify market basket analysis (association rule mining) for e-commerce operators, data analysts, and modern AI coding agents. By executing a single local CLI command, users can transform raw order data from Excel spreadsheets into actionable product pairing insights using the classic Apriori algorithm. Because the entire execution lifecycle occurs locally, data security is guaranteed as none of your confidential transaction records ever leave your machine.
As part of the broader ecosystem of Openclaw Skills, pairpulse provides structured JSON outputs that enable autonomous AI agents to parse, interpret, and act upon transaction insights. It dramatically cuts down analysis time from hours of manual spreadsheet manipulation to a simple 10-second command-line action, generating ready-to-use business strategy suggestions such as bundle package configurations, recommendation engine optimizations, and retail shelf adjustments.
pairpulse Use Cases
- E-Commerce Recommendation Optimization: Identify 'frequently bought together' patterns to strategically place cross-sell and up-sell suggestions across product detail pages.
- Bundle & Combo Design: Extract insights on top product combinations to create high-converting promotional bundles and discount schemes.
- Offline Retail Merchandising: Reorganize store layout and physical shelf arrangements based on scientific buyer affinity models.
- Autonomous AI Business Analysis: Equip LLM-driven agents with the ability to programmatically run local transactional audits and deliver automated strategic business advice within Openclaw Skills workflows.
How pairpulse Works
- Data Preparation: The user loads transaction data containing at least order IDs and product names from an Excel file.
- Product Name Standardization: Runs an optional standardization pass to fuzzy-match and unify disparate product names.
- Association Mining: Applies the Apriori algorithm to identify co-purchase patterns across customer baskets, calculating support, confidence, and lift.
- Strategy Mapping: Matches discovered association rules to specific operational strategies (e.g., bundling, up-selling).
- Multi-Format Export: Generates reports in Excel, visual PNG charts, or structured JSON designed for agent consumption.
pairpulse Setup
To integrate pairpulse into your workflows or run it with autonomous agents compatible with Openclaw Skills, start by verifying your installation or running a quick setup.
Verify Installation
Ensure the command is accessible on your system path:
pairpulse --version
Install via pip
If the command is not found, install it using Python's package manager:
pip install pairpulse
Upgrade to the Latest Version
To make sure your local AI agent is using the newest features, keep the tool updated:
pip install --upgrade pairpulse
pairpulse Data Schema & Taxonomy
Input File Requirements
The input dataset must be an Excel file (.xlsx) structured such that each row represents a single item inside a specific order.
| Column Name | Status | Description |
|---|---|---|
| Order ID (订单编号) | Required | Identifies unique transactions. |
| Product Name (商品名称) | Required | The label or SKU of the purchased item. |
| Unit Price (商品单价) | Recommended | Enables value-based cohort clustering. |
| Customer ID (客户ID) | Recommended | Unlocks repeat purchase and customer-level insights. |
| Product Category (商品类目) | Recommended | Allows category-level affinity analysis. |
Output Schema
When running with the --format json option, the structured payload includes:
summary: High-level metrics such as total rows parsed, total unique orders, and rules identified.rules_by_group: Association rules grouped by segment (e.g., high-value orders), highlightingantecedent(what they bought),consequent(what they also bought), and metrics (support,confidence,lift) alongside suggested marketing actions.insights: Natural language insights summarizing key takeaway trends.errors/warnings: Diagnostic details for missing fields or anomalies.
pairpulse Advanced Features
- Data Standardization Engine: Clean messy, non-uniform product naming conventions instantly using fuzzy standardization maps before running calculations.
- Segmented Cohort Analysis: Filter and run association mining by specific customer groups, order values, or item counts.
- Agent-Ready JSON Outputs: Designed explicitly to serve as structured computational outputs for tool-calling AI agents, elevating Openclaw Skills capabilities in automated retail analytics.
- Local-First & Offline Secure: Runs entirely on the host machine without external API dependencies, keeping sensitive transaction data confidential.
- Automatic Sync: Real-time integration with GitHub and PyPI allows agents and development systems to fetch the latest metadata and execution updates seamlessly.
SKILL.md
Loading
Related Openclaw Skills
MetricHub Multi-Tenant Metrics Platform
MetricHub is a multi-tenant SaaS metrics platform featuring query building, ECharts visualization, and dashboard orchestration optimized for Openclaw Skills.
belief-assessor
An LLM-driven epistemic reasoning engine that evaluates claims against retrieved evidence, outputting a structured belief state with calibrated confidence levels.
DeepRead Tax Forms
An AI-powered document extraction skill to convert W-2s, 1099s, 1040s, and scanned tax forms into clean, typed JSON with per-field confidence flags.
Alibaba Cloud DNS Resolution Diagnostics
An automated diagnostic utility designed to troubleshoot DNS resolution anomalies, domain unreachable issues, and routing misconfigurations across Alibaba Cloud and third-party DNS platforms.
Cloudflare Turnstile Spin Integration
An automated workflow skill that enables AI agents to configure Cloudflare Turnstile widgets, deploy verification Workers, and implement frontend bot protection end-to-end.
Cloudflare Durable Objects
A high-performance AI coding assistant skill designed to build, optimize, and test stateful, coordinated applications using Cloudflare Durable Objects.
Featured*
