스킬 azure-ai-formrecognizer-java

📦

azure-ai-formrecognizer-java

Name: azure-ai-formrecognizer-java
Author: sickn33

안전

Extract Data from Documents with Azure Form Recognizer

Automate document processing by extracting text, tables, and structured data from forms, invoices, and receipts. Build Java applications that leverage Azure Document Intelligence prebuilt models or create custom document analyzers.

지원: Claude Codex Code(CC)

🥉 74 브론즈

스킬 ZIP 다운로드

Claude에서 업로드

설정 → 기능 → 스킬 → 스킬 업로드로 이동

토글을 켜고 사용 시작

테스트해 보기

"azure-ai-formrecognizer-java" 사용 중입니다. Analyze receipt from restaurant image

예상 결과:

Extracted fields: Merchant='The Grand Bistro', TransactionDate=2024-01-15, Total=87.42, Subtotal=78.50, Tax=8.92. Line items: Pasta Carbonara ($24.00), Caesar Salad ($18.00), Tiramisu ($12.00), Wine ($24.50). Confidence scores above 0.95 for all fields.

"azure-ai-formrecognizer-java" 사용 중입니다. Extract table from quarterly report PDF

예상 결과:

Table detected: 12 rows x 5 columns. Headers: Quarter, Revenue, Expenses, Profit, Growth. Data extracted with cell positions and confidence scores. Table structure preserved with row/column indices for spreadsheet export.

"azure-ai-formrecognizer-java" 사용 중입니다. Process batch of mixed document types

예상 결과:

Document 1: Classified as Invoice (98% confidence) - Vendor: ABC Supply, Amount: $1,250.00. Document 2: Classified as Receipt (96% confidence) - Merchant: Office Depot, Total: $89.99. Document 3: Classified as Contract (94% confidence) - 8 pages detected.

보안 감사

안전

v1 • 2/24/2026

All static analysis findings are false positives. The file is Markdown documentation containing Java SDK code examples. No executable code, command injection, or malicious patterns detected. External command detections confused markdown code fences with shell syntax. URL references are documentation examples for Azure endpoint configuration.

스캔된 파일

347

분석된 줄 수

발견 사항

총 감사 수

보안 문제를 찾지 못했습니다

감사자: claude

품질 점수

아키텍처

100

유지보수성

콘텐츠

커뮤니티

100

보안

사양 준수

만들 수 있는 것

Automated Invoice Processing

Extract vendor names, invoice dates, line items, and totals from supplier invoices. Integrate with accounts payable systems to automate data entry and reduce manual processing time.

Receipt Data Capture

Build expense reporting applications that extract merchant names, transaction dates, and itemized purchases from receipt images. Enable mobile receipt scanning for employee reimbursements.

Custom Form Digitization

Create custom models for organization-specific forms like purchase orders, contracts, or survey responses. Train the model on sample documents and deploy automated extraction workflows.

이 프롬프트를 사용해 보세요

Quick Start: Analyze a Document

Create a Java program using Azure Document Intelligence SDK to extract text and tables from a PDF file. Use the prebuilt-layout model and print the results to console.

Receipt Analysis with Field Extraction

Write Java code to analyze a receipt image using the prebuilt-receipt model. Extract merchant name, transaction date, and total amount with confidence scores. Handle cases where fields may be missing.

Custom Model for Invoice Processing

Generate Java code to build a custom document model for invoice extraction. Include training from Azure Blob Storage, model composition from multiple invoice templates, and error handling for training failures.

Document Classification Pipeline

Create a complete Java solution that classifies incoming documents as invoices, receipts, or contracts using a custom classifier, then routes each type to the appropriate prebuilt model for detailed extraction. Include async polling and retry logic.

모범 사례

Use DefaultAzureCredential for production deployments to leverage managed identities and avoid hardcoding credentials
Implement polling with appropriate timeouts for long-running analysis operations on large documents
Validate document format and size before submission to avoid unnecessary API calls and costs

피하기

Do not hardcode API keys or endpoints in source code; use environment variables or Azure Key Vault
Avoid synchronous blocking calls in high-throughput scenarios; use async patterns for scalability
Do not process documents without validating file types; unsupported formats will fail analysis

자주 묻는 질문

What document formats does Azure Form Recognizer support?

Azure Document Intelligence supports PDF, JPEG, PNG, BMP, and TIFF formats. PDF documents can be up to 500 pages. Maximum file size is 500 MB for PDF and 20 MB for images.

How accurate are the prebuilt models?

Prebuilt models typically achieve 95%+ accuracy on standard documents. Accuracy varies by document quality, language, and field type. Custom models trained on your specific documents can achieve higher accuracy for specialized forms.

What is the difference between prebuilt and custom models?

Prebuilt models work out-of-box for common document types like invoices and receipts. Custom models are trained on your specific document templates for fields not covered by prebuilt models, such as proprietary forms or specialized layouts.

How do I handle documents in multiple languages?

Azure Document Intelligence supports over 70 languages automatically. Prebuilt models detect language from document content. For custom models, include training samples in all target languages for best results.

What happens if the analysis operation fails?

Failed operations return an error with status code and message. Common errors include invalid document format, unsupported language, or service throttling. Implement retry logic with exponential backoff for transient failures.

Can I process documents stored in Azure Blob Storage directly?

Yes, you can pass a SAS URL to beginAnalyzeDocumentFromUrl instead of uploading file bytes. This is more efficient for large documents and reduces bandwidth usage. Ensure the SAS token has read permissions and adequate validity period.

개발자 세부 정보

작성자

sickn33

라이선스

MIT

리포지토리

https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/azure-ai-formrecognizer-java

참조

main

파일 구조

📄 SKILL.md