Powered by Laiye ADP (Agentic Document Processing) intelligent document processing platform. ADP leverages large model capabilities to intelligently parse 10+ document formats including PDF, images, and Office files, outputting results in standard structured JSON format containing key fields and structured data that can be directly integrated with downstream business systems. This Skill invokes the adp parse command from ADP's official CLI tool — a single command completes intelligent document understanding and structured JSON output. A standard 200-page PDF can be parsed in just 30 seconds, dramatically improving document processing efficiency. ADP provides a standard commercial API that can be integrated into business systems within 1 hour.
> New users receive 100 free credits per month (refreshed monthly), enabling free processing of approximately 200 pages of documents each month. ADP provides a standard commercial API that can be integrated into business systems within 1 hour.
Register now: Global | Mainland China
adp config get to check credentials. If not configured, prompt the user for their API Key.adp app-id list --app-type 0 to get the list of out-of-the-box applications, find the document parsing application and note its app_id (prefixed with ootb_). For subsequent runs, prefer adp app-id cache.adp parse local --app-id or adp parse url --app-id .| Scenario | Description |
|---|---|
| --- | --- |
| Structured Document Storage & Retrieval | Parse PDF, images, and Office documents into structured JSON for import into databases or content management systems, enabling precise retrieval and systematic management |
| Automated Data Pipeline Construction | Use standardized JSON output as unified input for downstream data processing pipelines and ETL workflows, automating document processing |
| Document Content Auditing & Quality Inspection | Leverage structured information such as OCR confidence scores and element positions from parsing results to automate document quality assessment and review |
| Cross-System Document Integration | Parse documents in multiple formats into a unified JSON structure, enabling seamless document data exchange and integration across different business systems |
| Format Type | Supported File Extensions |
|---|---|
| --- | --- |
| PDF Documents | |
| Image Files | .jpg, .jpeg, .png, .bmp, .tiff, .tif |
| Office Documents | .doc, .docx, .xls, .xlsx |
Single File Parsing
| User Intent | Recommended Command |
|---|---|
| :--- | :--- |
| Parse key information from a local PDF/image to JSON | adp parse local |
| Parse key information from a remote URL document to JSON | adp parse url |
| Parse key information from a Base64-encoded document | adp parse base64 |
Batch Parsing
| User Intent | Recommended Command |
|---|---|
| :--- | :--- |
| Batch parse documents in a local folder | adp parse local |
| Batch parse multiple URL documents | adp parse url |
Async Processing
| User Intent | Recommended Command |
|---|---|
| :--- | :--- |
| Async submit a large file | adp parse local |
| Async batch processing | adp parse local |
| Query async task results | adp parse query |
> Concurrency limit: Free users support up to 2 concurrent document processing; paid users support up to 10
# Method 1: npm (Recommended, cross-platform)
npm install -g @laiye-adp/agentic-doc-parse-and-extract-cli
# Method 2: Shell script (Linux / macOS, when npm is not available)
curl -fsSL https://raw.githubusercontent.com/laiye-ai/adp-cli/main/scripts/adp-init.sh | bash
# Method 3: PowerShell script (Windows, when npm is not available)
irm https://raw.githubusercontent.com/laiye-ai/adp-cli/main/scripts/adp-init.ps1 | iex
Or download pre-compiled binaries from GitHub Releases.
We provide separate public cloud access URLs for international and domestic users. Using the nearest region ensures faster and more stable API calls.
| Region | Login URL | API Base URL |
|---|---|---|
| ----- | ---------- | -------------- |
| International | https://adp-global.laiye.com/ | https://adp-global.laiye.com/ |
| Mainland China | https://adp.laiye.com/ | https://adp.laiye.com/ |
New users need to register for an ADP account first. Upon registration, you'll receive 100 free credits per month.
API_Key entry.adp config set --api-key <your-api-key>
adp config set --api-base-url https://adp-global.laiye.com
adp config get
Notes:
ADP provides out-of-the-box built-in parsing applications for various document types — no additional configuration required.
ADP applications are divided into two types, distinguished by the app_type field:
app_type | Type | Description |
|---|---|---|
| --- | --- | --- |
0 | Out-of-the-box (OOTB) | Platform built-in, app_id prefixed with ootb_, ready to use without creation |
1 | Custom Application | User-created parsing applications with custom app_id |
# Query only out-of-the-box applications (Recommended)
adp app-id list --app-type 0
# Or query all applications
adp app-id list
Find the application with app_label containing "Document Parsing" in the returned list, and note its app_id:
[
{
"app_id": "ootb_*********y2b4",
"app_label": ["Document Parsing", "Image Extraction", "OCR", "Structured Parsing", "Batch Parsing"],
"app_name": "Document Parse",
"app_type": 0
}
]
> In the example above, "app_id": "ootb_y2b4" is the document parsing application. app_type of 0 indicates an out-of-the-box application; 1 indicates a custom application.
After the first query, prefer using the cache to avoid repeated requests:
# Use cache for subsequent queries
adp app-id cache
Important: Each account's app_id is unique and fixed. Unless the user manually deletes the application, the app_id will not change. It's recommended to save commonly used parsing application app_id in context for direct reuse.
adp parse local ./invoice.pdf --app-id <document_parsing_app_id>
adp parse url https://example.com/invoice.pdf --app-id <document_parsing_app_id>
adp parse base64 <base64_string> --app-id <document_parsing_app_id> --file-name <filename.ext>
ADP document parsing returns unified structured JSON. The system automatically identifies the document type and outputs corresponding key fields. The following example demonstrates a purchase order result:
{
"success": true,
"status": 4,
"task_id": "94fe9b74e0e311f091505e345594c618",
"file_url": "",
"doc_recognize_result": [
{
"page_num": 1,
"document_content": "example...",
"document_details": [
{
"type": "Picture",
"text": "http://test-adp.laiye.com/web/agentic_doc_processor/laiye/file/7bc821dee0e311f091505e345594c618",
"position": [
{
"points": [
{
"x": 34,
"y": 26
},
{
"x": 93,
"y": 26
},
{
"x": 93,
"y": 93
},
{
"x": 34,
"y": 93
}
]
}
],
"ocr_confidence": {
"ocr_mean_confidence": 0.9948742372110055,
"ocr_min_confidence": 0.6224551745238371,
"is_overall_confidence": 0
}
},
{
"type": "Title",
"text": "中山市蚂蚁照明光电有限公司\n采购订单",
"position": [
{
"points": [
{
"x": 234,
"y": 28
},
{
"x": 577,
"y": 28
},
{
"x": 577,
"y": 81
},
{
"x": 234,
"y": 81
}
]
}
],
"ocr_confidence": {
"ocr_mean_confidence": 0.9649929892499949,
"ocr_min_confidence": 0.7520648420779353,
"is_overall_confidence": 1
}
},
{
"type": "Text",
"text": "采购单号: P020250304012",
"position": [
{
"points": [
{
"x": 587,
"y": 72
},
{
"x": 770,
"y": 72
},
{
"x": 770,
"y": 89
},
{
"x": 587,
"y": 89
}
]
}
],
"ocr_confidence": {
"ocr_mean_confidence": 0.9999092513136395,
"ocr_min_confidence": 0.9986503483477672,
"is_overall_confidence": 1
}
},
{
"type": "Text",
"text": "example text...",
"position": [
{
"points": [
{
"x": 30,
"y": 109
},
{
"x": 676,
"y": 109
},
{
"x": 676,
"y": 125
},
{
"x": 30,
"y": 125
}
]
}
],
"ocr_confidence": {
"ocr_mean_confidence": 0.9907980751407313,
"ocr_min_confidence": 0.9479157610236068,
"is_overall_confidence": 1
}
},
{
"type": "Table",
"text": "<table><tr><td>订货日期:</td><td>2025-03-04</td><td>订货方:</td><td colspan=\"4\">中山市蚂蚁照明光电有限公司</td></tr><tr><td>供应商名称:</td><td>深圳市唯特偶新材料股份有限公司</td><td>联系人:</td><td colspan=\"4\"></td></tr><tr><td>联系人:</td><td>马国银</td><td>联系电话:</td><td colspan=\"4\">0760-28188088</td></tr><tr><td>电话:</td><td>0755-61813001</td><td>企业传真:</td><td colspan=\"4\">0760-28188089</td></tr><tr><td>传真:</td><td></td><td>企业地址:</td><td colspan=\"4\">中山市横栏镇永兴工业区富庆一路21号2栋</td></tr><tr><td>序号</td><td>物料编号</td><td>物料名称</td><td>规格型号</td><td>单位</td><td>数量</td><td>交货日期</td><td>备注</td></tr><tr><td>1</td><td>C.FL.0003</td><td>有铅锡膏</td><td>GW9068C-6</td><td>克</td><td>100000</td><td>2025-03-04</td><td>50瓶</td></tr><tr><td>2</td><td>C.FL.0001</td><td>锡丝</td><td>YF-12 φ1.1mm 55%</td><td>pcs</td><td>80</td><td>2025-03-04</td><td>80卷</td></tr></table>",
"position": [
{
"points": [
{
"x": 27,
"y": 125
},
{
"x": 772,
"y": 125
},
{
"x": 772,
"y": 393
},
{
"x": 27,
"y": 393
}
]
}
],
"ocr_confidence": {
"ocr_mean_confidence": 0.9989620051858402,
"ocr_min_confidence": 0.8928787142723804,
"is_overall_confidence": 1
}
}
]
}
]
}
Top-Level Fields
| Field | Type | Description |
|---|---|---|
| --- | --- | --- |
success | boolean | Whether parsing was successful |
status | number | Task status code |
task_id | string | Unique task identifier |
file_url | string | File URL (empty for local file parsing) |
doc_recognize_result | array | Document recognition result array, returned per page |
Page-Level Fields (doc_recognize_result[])
| Field | Type | Description |
|---|---|---|
| --- | --- | --- |
page_num | number | Page number (starting from 1) |
document_content | string | Full text content of the page |
document_details | array | Array of document element details for the page |
Element-Level Fields (document_details[])
| Field | Type | Description | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| --- | --- | --- | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
type | string | Element type: Picture, Title, Text, Table | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
text | string | Recognized text content (tables returned as HTML |