Identify and cluster the root causes of product returns to drive actionable prevention strategies and preserve business margins.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install refund-reason-cluster
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install refund-reason-cluster using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
The Refund Reason Cluster skill is a specialized analytical tool designed for AI agents to parse through messy refund logs, return reason text, and support transcripts. By leveraging Openclaw Skills, developers and e-commerce operators can transform raw post-purchase data into structured clusters categorized by quality, fit, shipping, or expectation gaps. This skill goes beyond simple categorization by providing root-cause hypotheses and identifying which refunds are avoidable versus unavoidable.
This utility is essential for businesses looking to minimize margin erosion caused by returns. It synthesizes disparate data points into an actionable executive summary, allowing teams to prioritize high-impact product fixes and update pre-purchase marketing copy to better align with customer expectations.
To integrate this skill into your AI agent workflow using Openclaw Skills, ensure your environment has access to order metadata and refund logs. Use the following commands to get started:
# Import the skill into your Openclaw environment
openclaw add refund-reason-cluster
# Configure data access for refund logs and support transcripts
openclaw config set REFUND_DATA_PATH="./data/returns.csv"
The skill organizes data into a structured hierarchy to ensure clarity between signals and actions:
| Component | Description | Examples |
|---|---|---|
| Input Source | Raw unstructured and structured logs | Refund logs, Return text, Support transcripts |
| Cause Clusters | High-level grouping of issues | Quality, Fit, Shipping, Expectation, Misuse |
| Evidence Table | Validation metrics for each cluster | Signal strength, Confidence score, Financial risk |
| Execution Plan | Time-bound recovery steps | Short-term fixes (7 days), Long-term product roadmap |
Loading
A comprehensive toolkit of pre-installed Python and C++ examples for AI inference, hardware control, and multimedia processing on the RDK X5 platform.

A strategic scheduling tool that balances ecommerce promotion intensity with operational capacity and inventory constraints.

A Windows-optimized companion skill that bridges local AI CLIs with secure, controlled workspace file operations.

A sophisticated coordination layer for multi-agent workflows featuring shared task boards, message inboxes, and lease-based task management.

An AI-powered side hustle consultant that scans freelance platforms, analyzes project feasibility, and recommends the best moneymaking opportunities based on your tech stack.

A professional-grade batch processing utility that converts legacy and modern Office documents into structured Markdown while preserving formatting and directory hierarchy.








































