A privacy-first financial auditor that scans bank statements to identify recurring charges, flag suspicious transactions, and generate refund request templates.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install refund-radar
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 refund-radar 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 Refund Radar?
Refund Radar is a sophisticated local utility designed to help users regain control over their finances by automating the tedious process of statement auditing. As part of the Openclaw Skills ecosystem, it processes bank and credit card CSV exports or raw text locally on your machine, ensuring complete privacy. It goes beyond simple tracking by identifying recurring cadences, flagging duplicate charges, and noting significant price spikes in your subscriptions.
The tool synthesizes complex financial data into a user-friendly, interactive HTML report. This report includes actionable insights and ready-to-copy refund templates for different platforms like email or live chat. By utilizing Openclaw Skills like this one, developers and power users can manage their financial hygiene without relying on third-party cloud aggregators or sharing sensitive banking credentials.
Refund Radar Use Cases
- Scanning monthly bank statements to uncover forgotten or orphaned subscriptions.
- Detecting duplicate transactions and accidental double-billing by merchants.
- Identifying stealthy price increases and unexpected service fees.
- Generating formal refund requests for unauthorized or suspicious charges.
- Performing a privacy-conscious annual audit of personal or business spending habits.
How Refund Radar Works
- Transaction data is ingested from CSV exports (Chase, Apple Card, etc.) or pasted text.
- The system normalizes the data, auto-detecting delimiters, date formats, and currencies.
- Heuristic analysis identifies recurring charges based on frequency, merchant keywords, and amount consistency.
- High-severity flags are triggered for duplicates, amount spikes, and unusual currency anomalies.
- The user reviews flagged items via CLI to update the local merchant knowledge base.
- A rich, interactive HTML report is generated with privacy toggles and collapsible sections for easy viewing.
- Customizable refund templates are drafted in multiple tones (Concise, Firm, Friendly) for immediate use.
Refund Radar Setup
To get started with this entry in the Openclaw Skills library, ensure you have Python 3.9+ installed. No external dependencies are required.
# Analyze a standard CSV export
python -m refund_radar analyze --csv statement.csv --month 2026-01
# Analyze from pasted text via stdin
python -m refund_radar analyze --stdin --month 2026-01 --default-currency USD
Refund Radar Data Schema & Taxonomy
Refund Radar organizes its logic and output into a local directory structure for persistence and auditing:
| Path | Purpose |
|---|---|
~/.refund_radar/state.json |
Stores learned merchant history and user preferences. |
~/.refund_radar/reports/YYYY-MM.html |
The primary interactive visual audit report. |
~/.refund_radar/reports/YYYY-MM.json |
Raw structured data of the monthly analysis results. |
Refund Radar Advanced Features
- Local-first architecture with zero external API calls for maximum data security.
- Interactive HTML reports featuring a privacy toggle to blur merchant names during screen sharing.
- Multi-tone refund request generation for email, chat, and bank dispute forms.
- Learning engine that remembers expected merchants to reduce false positives over time.
- Intelligent detection of Fee-like charges such as ATM, overdraft, and service fees.
SKILL.md
Loading
Related Openclaw Skills
MetaMask Agent Wallet
A secure, sandboxed MetaMask wallet interface for autonomous AI agents to execute blockchain transactions within user-defined safety limits.
Todoist CLI Manager
A robust CLI tool for managing Todoist tasks, projects, and labels via the REST API v2.
Just Calendar CLI
A terminal-based interface for managing calendar data and habit tracking stored directly in Google Drive via the justcalendar.ai platform.
Cubistic Public Bot Integration
A technical framework for onboarding and managing external bots within the Cubistic 3D ecosystem using Proof-of-Work verification.
WalletPilot
WalletPilot provides a secure framework for AI agents to automate interactions with over 10 major crypto wallets across EVM and Solana chains.
WalletPilot-7715
A secure framework for AI agents to execute on-chain transactions using MetaMask ERC-7715 permissions without private key exposure.
Featured*
