A sophisticated meta-skill that automates the pre-accounting lifecycle by orchestrating email intake, OCR data extraction, payment verification, and accounting record creation.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install bookkeeper
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 bookkeeper 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 Autonomous Bookkeeper?
The Autonomous Bookkeeper is a powerful orchestration tool designed to bridge the gap between incoming financial documents and your accounting software. By leveraging Openclaw Skills, this agent manages the complex logic required to detect invoices in your inbox, extract high-fidelity data, and reconcile those documents against actual payment events. It acts as a digital clerk that ensures your records are accurate, reconciled, and ready for review without the need for manual data entry.
Built on a foundation of modularity, this skill integrates several specialized Openclaw Skills to handle niche tasks. It provides a robust framework for financial automation, allowing businesses to maintain strict financial controls while significantly reducing the administrative overhead associated with accounts payable and expense management.
Autonomous Bookkeeper Use Cases
- Automating the intake of recurring SaaS invoices like AWS or Google Cloud and mapping them to specific expense categories.
- Streamlining accounts payable by automatically creating reconciliation-ready bills in Xero from Gmail attachments.
- Reducing manual verification time by cross-referencing OCR-extracted invoice totals with Stripe transaction records.
- Implementing a high-confidence auto-posting workflow for trusted vendors while flagging edge cases for manual review.
How Autonomous Bookkeeper Works
- The agent scans Gmail for candidate emails containing attachments and invoice-related keywords.
- Detected attachments are processed through DeepRead OCR to extract critical fields such as vendor name, invoice date, and total amount.
- The system queries Stripe to verify if a matching payment exists within a defined date and amount tolerance window.
- A normalized data contract is generated, ensuring all metadata from the source email and the payment gateway is aligned.
- The agent creates an accounting entry in Xero (ACCPAY) and marks it as paid or reconciled based on the verification confidence.
- Traceability labels and logs are applied to the source email to maintain a clear audit trail and prevent duplicate processing.
Autonomous Bookkeeper Setup
To deploy this automation, you must first install the required sub-skills from the Openclaw Skills ecosystem and configure your API credentials.
# Install the necessary underlying skills
npx -y clawhub@latest install gmail deepread-ocr stripe-api xero
# Set required environment variables
export MATON_API_KEY="your_maton_api_key"
export DEEPREAD_API_KEY="your_deepread_api_key"
Autonomous Bookkeeper Data Schema & Taxonomy
The skill utilizes a normalized JSON schema to ensure data consistency across different platforms. Below is the organization of the transaction metadata:
| Object | Key Fields |
|---|---|
| Source | gmail_message_id, sender, attachment_name |
| Invoice | vendor, invoice_number, total, tax, currency, ocr_confidence_ok |
| Payment | provider (stripe), matched (boolean), transaction_id, amount |
| Accounting | system (xero), entry_type (ACCPAY), category, status |
Autonomous Bookkeeper Advanced Features
- Intelligent vendor-based routing rules to automatically assign expense categories based on the sender's identity.
- Multi-stage quality gates that prevent auto-posting of financial records if OCR confidence is low or if Human-in-the-Loop (HIL) flags are triggered.
- Customizable date and amount tolerance settings to accommodate transaction delays and minor currency discrepancies.
- Idempotency key management to ensure that a single invoice is never booked twice in the accounting system.
- Audit trail preservation by linking source Gmail message IDs directly to Xero reference fields.
SKILL.md
Loading
Related Openclaw Skills
Qwen3-TTS MLX
A high-performance local text-to-speech engine optimized for Apple Silicon that supports voice cloning and natural language voice design.
libvips Image Processing Skill
A high-performance image processing skill utilizing libvips for memory-efficient resizing, format conversion, and batch operations.
Gemini Watermark Remover
An offline Python tool that uses reverse alpha blending to mathematically remove the star logo from Gemini-generated images.
Garmin Health Report
A powerful Python-based tool for generating comprehensive daily health and professional running reports from Garmin Connect data.
Cold Outreach Hunter
An advanced orchestration meta-skill that automates the end-to-end B2B cold outreach pipeline from lead sourcing to personalized sequence generation.
Human Masked Content Creator
A sophisticated meta-skill that orchestrates multiple specialized agents to produce authentic, high-signal content that avoids robotic AI patterns.
METADATA
Featured*
