MOVA Invoice OCR & Approval for Openclaw

A professional-grade OCR and human-in-the-loop workflow for processing invoices, receipts, and financial documents with cryptographic audit trails.

mova-compact
v1.0.2
Apr 2, 2026
0
497
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install mova-invoice-ocr

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install mova-invoice-ocr 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 MOVA Invoice OCR & Approval?

The MOVA Invoice OCR skill is a robust solution within the Openclaw Skills ecosystem designed for secure financial document processing. It leverages MOVA HITL (Human-in-the-Loop) technology to automate the extraction of critical data from invoices and bills while ensuring that a human gatekeeper remains in control of the final decision. This approach combines the speed of AI with the accountability required for enterprise financial operations.

Every action taken through this skill is recorded in an immutable, cryptographically signed compact journal. This ensures that all OCR extractions, risk assessments, and human approvals are transparent and verifiable, making it an ideal tool for organizations needing to comply with high-risk financial regulations and human oversight requirements like the EU AI Act.

MOVA Invoice OCR & Approval Use Cases

  • Automating the extraction of line items, VAT, and IBAN data from supplier invoices.
  • Implementing a mandatory human approval gate for high-value financial transactions.
  • Performing risk validation to detect duplicate invoices or unauthorized vendor changes.
  • Generating tamper-proof audit trails for internal compliance and external financial audits.

How MOVA Invoice OCR & Approval Works

  1. The user provides a document image URL and triggers the workflow with a request to process or review the invoice.
  2. The skill utilizes the mova_hitl_start tool to perform automated OCR extraction and cross-reference data against ERP systems.
  3. A risk score and anomaly flags (e.g., VAT mismatch, unknown vendor) are generated and presented to the user.
  4. The process enters a 'waiting_human' state, requiring the user to select a decision: approve, reject, escalate, or request info.
  5. Upon decision, the skill executes the mova_hitl_decide tool to finalize the contract and stores the signed reasoning.
  6. An immutable audit receipt and compact journal are generated to provide a permanent record of the entire event chain.

MOVA Invoice OCR & Approval Setup

To integrate this capability into your workspace, you must install the MOVA plugin for Openclaw Skills using the following command:

openclaw plugins install openclaw-mova

Ensure you have access to the api.mova-lab.eu endpoint for document processing. For production environments, you can register your own ERP or OCR connectors using the mova_register_connector tool.

MOVA Invoice OCR & Approval Data Schema & Taxonomy

The skill organizes extracted metadata and audit events into a structured taxonomy:

Field Group Details
Extraction Vendor, IBAN/BIC, Tax ID, Line Items, Totals, VAT, OCR Confidence
Validation Risk Score (0.0-1.0), Anomaly Flags (iban_change, duplicate, vat_mismatch)
Compliance Audit Receipt ID, Compact Journal (signed event log), Human Reasoning
References PO Match Reference, Contract ID (unique session identifier)

MOVA Invoice OCR & Approval Advanced Features

  • Multi-connector support for linking real-world ERP and Vision LLM systems.
  • Cryptographic proof of decision-making for immutable record keeping.
  • Automated policy enforcement that routes anomalies to human review by default.
  • Support for custom internal hooks via the MOVA plugin architecture.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*