A smart AI-driven bookkeeping assistant that parses natural language inputs to automatically organize, log, and visualize your daily expenses locally.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install auto-bookkeeping
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 auto-bookkeeping 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 Auto Bookkeeping Assistant?
The Auto Bookkeeping skill is a powerful local utility designed to make personal finance management effortless. Instead of manually filling out tedious expense forms, you can log transactions using natural language (e.g., "Yesterday lunch was 30 dollars"). The underlying Python parsers automatically extract dates, currencies, transaction types, and categories, translating human speech into structured data. By extending your command-line environment, this skill seamlessly fits into developer workflows as one of many highly productive Openclaw Skills.
All transactional data is strictly stored locally using a secure SQLite database, ensuring absolute privacy. Users can generate rich, interactive HTML monthly reports featuring daily spending trends and categorical pie charts, making financial tracking visual and actionable without exposing sensitive data to third-party web services.
Auto Bookkeeping Assistant Use Cases
- Logging daily expenses instantly using simple, conversational natural language phrasing.
- Consolidating financial habits without manual database entry or complex SaaS solutions.
- Generating visual monthly summaries and categorical spending pie charts.
- Quickly retrieving past transaction history using targeted semantic searches.
- Keeping bookkeeping records entirely private and local within a secure SQLite database.
How Auto Bookkeeping Assistant Works
- Initialize: The skill verifies or initializes the local SQLite database automatically.
- Parse Input: The user provides a natural language entry which is processed by the internal Python parsing script to extract dates, amounts, categories, and confidence scores.
- Human-in-the-Loop Confirmation: If confidence falls below 0.6 or vital parameters like the amount are missing, the skill asks for user confirmation.
- Save Transaction: Once validated, the JSON representation of the transaction is written securely to the local SQLite database.
- Report & Visualize: When queried, the skill aggregates monthly metrics and compiles them into a clean HTML dashboard with visualization graphs.
Auto Bookkeeping Assistant Setup
Prerequisites
- Python environment
- Bash/Zsh shell access
Database Initialization
Run the initialization command before your first use (though it will automatically trigger on demand):
python {baseDir}/scripts/bookkeeper.py init
Usage Workflow Examples
1. Parsing User Input
python {baseDir}/scripts/parser.py "Yesterday lunch 30"
2. Recording an Expense
python {baseDir}/scripts/bookkeeper.py add '{"date":"2026-06-13","amount":30.0,"category":"Dining","note":"Lunch","type":"expense"}'
3. Generating HTML Visual Report
# Get monthly summary metrics
SUMMARY=$(python {baseDir}/scripts/bookkeeper.py summary 2026 06)
# Retrieve recent records
ENTRIES=$(python {baseDir}/scripts/bookkeeper.py list 2026 06 100 0)
# Compile into an interactive dashboard
python {baseDir}/scripts/reporter.py "$SUMMARY" "$ENTRIES"
Auto Bookkeeping Assistant Data Schema & Taxonomy
File Organization
- Database File:
{baseDir}/data/bookkeeping.db(Local SQLite database) - Monthly Visual Reports:
{baseDir}/data/report_<year>_<month>.html
Data Schema (Transaction JSON payload)
| Field | Type | Description |
|---|---|---|
date |
String | Format YYYY-MM-DD, extracted from query |
time |
String | Format YYYY-MM-DD HH:MM:SS, default timestamp |
amount |
Float | Value of transaction |
category |
String | Normalized transaction category |
note |
String | Contextual memo details |
type |
String | expense (支出) or income (收入) |
confidence |
Float | AI parser confidence score (0.0 to 1.0) |
raw |
String | Original raw natural language string |
parsed_ok |
Boolean | True if parsing completed successfully |
Categorization Taxonomy
| Category | Trigger Keyword Examples |
|---|---|
| Food & Dining | Meal, lunch, milk tea, coffee, takeout |
| Transportation | Taxi, Uber, bus, subway, gas |
| Shopping | Supermarket, shopping, Amazon, Taobao |
| Entertainment | Movies, gaming, KTV, gym |
| Housing | Rent, water bill, electricity, property management |
| Medical | Hospital, medicine, health check |
| Salary | Base pay, bonus, wages |
| Side Hustle | Freelance, royalties, commission |
| Investments | Interest, dividends, funds |
Auto Bookkeeping Assistant Advanced Features
- Local-First Privacy: Zero external cloud reliance using SQLite and locally-compiled HTML templates.
- High-Confidence Direct Logging: Intelligent confidence scoring skips confirmation prompts for unambiguous records (confidence >= 0.6).
- Fuzzy Date Processing: Handles relative references such as "yesterday", "last Friday", "3 days ago", or "June 5th".
- Semantic Search: Retrieve historical logs using key phrases, specific items, or pricing criteria.
- Seamless Extensibility: Easily integrates as one of your personalized Openclaw Skills to connect with voice interfaces or chat agents.
SKILL.md
Loading
Related Openclaw Skills
AI Sports Health Assistant
A scientific, natural-language-driven workout logger and training planner that uses precise MET calorie estimations and visual HTML performance reporting.
Food Nutrition Assistant
An AI-powered nutrition consultant that simplifies meal logging and dietary analysis using natural language and the Chinese Dietary Guidelines (DRIs 2024).
ABN Business Verification
A live, read-only lookup skill to verify Australian businesses, ABNs, ACNs, and GST registration status directly via the official Australian Business Register.
WeChat Mini Program Development Feasibility Decision Assistant
Evaluate WeChat Mini Program concepts instantly with automated market research, WeChat Index analysis, and professional HTML feasibility reports.
Gumroad Admin
An advanced automation skill designed to administer Gumroad e-commerce stores, streamline batch digital product publishing, and safely manage files using the official CLI.
Financial Astrology AI Agent Skill
An advanced AI coding agent skill that analyzes market cycles, crowd psychology, and technical planetary alignments for automated day trading and financial research.
Featured*
