Veryfi Documents AI is a real-time OCR and data extraction engine designed to transform unstructured documents into structured, actionable JSON data.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install documents-ai
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 documents-ai 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 Veryfi Documents AI?
Veryfi Documents AI provides a sophisticated API-driven approach to document intelligence, specializing in the extraction of granular data from financial and legal documents. By utilizing this within the framework of Openclaw Skills, developers can grant AI agents the ability to read and understand receipts, invoices, bank statements, and tax forms with exceptional speed and accuracy.
Beyond simple text recognition, this skill offers advanced document classification, fraud detection, and line-item extraction. It serves as a critical bridge between physical or digital documents and automated workflows, ensuring that Openclaw Skills users can process high volumes of paperwork without manual intervention or data entry errors.
Veryfi Documents AI Use Cases
- Automating expense reporting by extracting line items and tax data from receipts.
- Streamlining accounts payable workflows through automated invoice processing.
- Analyzing financial health by parsing transaction histories from bank statements.
- Validating compliance for tax documents such as W-9, W-2, and W-8 forms.
- Routing documents to specific departments using automated document classification.
How Veryfi Documents AI Works
- Authentication is established using Veryfi Client ID, Username, and API Key credentials.
- The document is submitted to the API via multipart file upload, a public URL, or base64-encoded string.
- Veryfi's AI engine performs OCR and identifies the document's structure based on its type or a specific blueprint.
- The engine extracts key-value pairs, line items, and metadata while performing fraud checks.
- A structured JSON response is returned to the agent, providing everything from vendor details to raw OCR text.
Veryfi Documents AI Setup
First, obtain your API credentials from the Veryfi dashboard. Then, configure your environment variables to enable the skill for use with Openclaw Skills:
export VERYFI_CLIENT_ID="your_client_id"
export VERYFI_USERNAME="your_username"
export VERYFI_API_KEY="your_api_key"
Enable the skill in your configuration file:
{
"skills": {
"entries": {
"veryfi-documents-ai": {
"enabled": true
}
}
}
}
Veryfi Documents AI Data Schema & Taxonomy
The skill organizes extracted data into specialized schemas based on the document endpoint used. Below is a summary of the metadata taxonomy:
| Feature | Description |
|---|---|
| Vendor Info | Name, address, phone, and website of the document issuer. |
| Line Items | Detailed breakdown of products, quantities, prices, and taxes. |
| Financial Totals | Subtotal, tax lines, tip, discount, and grand total. |
| Fraud Metadata | Fraud scores, decision flags, and detection of AI-generated content. |
| OCR Text | A full plain-text string of all detected characters on the document. |
| MICR Data | Routing and account numbers extracted from checks. |
Veryfi Documents AI Advanced Features
- Custom Blueprints: Create tailored extraction logic for unique document types not covered by default endpoints.
- Bounding Box Coordinates: Retrieve exact positional data for layout analysis and visual verification.
- Multi-Page Support: Process complex bank statements up to 50 pages or invoices up to 15 pages.
- Fraud & Forensic Analysis: Identify tampered documents using color-coded fraud scores and decision logic.
- Flexible Input Methods: Support for multipart file uploads, base64 data, and direct URL processing for diverse agent environments.
SKILL.md
Loading
Related Openclaw Skills
Vector Control
A comprehensive interface for controlling Vector robots through Wirepod local HTTP APIs and CLI scripts.
Railway Deploy
Automate code deployments to Railway using the CLI with support for custom messages, service selection, and real-time log streaming.
Academic Writer & LaTeX Composer
A professional LaTeX writing assistant that automates the lifecycle of academic paper creation from template analysis to PDF compilation.
Beaver Habit Tracker
An AI-powered integration for the Beaver Habit Tracker API that automates habit logging and visualization through an intelligent agent interface.
PDF To Structured Data Extraction
Automate the extraction of structured data from complex construction PDFs, specifications, and Bill of Materials using advanced OCR and parsing.
Amap (Gaode Maps)
A powerful geospatial skill providing AI agents with real-time weather data, POI searching, and driving route planning via the Amap Web Service API.
METADATA
Featured*
