Extract Images from Article URL API
Extract ordered image candidates from an article
Bestseller
AI Extraction Structured JSON Automation
AI Structured Web Data Extraction API
Extract webpage fields with a prompt.
What can this API do?
Webpage fields from an instruction
Provide a webpage URL and extraction instruction, then receive structured JSON fields for the requested data.
AI agent extraction workflow
Use the response in agents, automations, and RAG pipelines that need clean structured data from public webpages.
Flexible structured results
Describe the fields you need, then validate the extracted JSON for your application. Explicit empty lists represent no matches.
Reduces custom parser work
Avoid writing one-off scrapers for every page layout when your product only needs a specific structured result.
Automation-ready response fields
Return JSON that backend jobs, no-code tools, enrichment workflows, and internal dashboards can consume directly.
Standalone structured extraction API
Subscribe to this interface when the output should be JSON; use fetchcontent when the workflow mainly needs article text.
Annual plan
$49$99
Try it for free!
Sign In Sign in to get a trial key and test all APIs.
Secure checkout powered by Stripe
API Document
HTTP Protocol:HTTPS
HTTP Method:POST
HTTP Endpoint:https://api.gugudata.io/v1/websitetools/url2json
Response Type:application/json; charset=utf-8
DEMO Endpoint:https://api.gugudata.io/v1/websitetools/url2json/demo
Live Demo:Try Interactive Demo
Full API Docs:developers.gugudata.io
API Request Parameters
| Name | Type | Is Required | Default Value | Remark |
|---|---|---|---|---|
| appkey | string | true | YOUR_APPKEY | Application key supplied as a query parameter. |
| url | string | true | Public HTTP or HTTPS webpage URL to extract from. | |
| prompt | string | true | Describe the fields to extract. Returned fields depend on your instruction and the page content. |
API Response Parameters
| Name | Type | Remark |
|---|---|---|
| dataStatus | object | Response metadata returned by the current v1 contract. |
| dataStatus.requestParameter | string | Page host, path and instruction length. The Demo shows its public sample URL. Private instructions, URL query values and credentials are omitted. |
| dataStatus.statusCode | integer | Application-level status code returned by the current v1 contract. Successful demo responses currently return `200`. |
| dataStatus.status | string | Application-level status enum. Successful demo responses currently return `SUCCESS`. |
| dataStatus.statusDescription | string | Application-level status message returned by the current v1 contract. |
| dataStatus.responseDateTime | string | Response timestamp returned by the current service contract. |
| dataStatus.dataTotalCount | integer | Total number of records that match the request. |
| data | object | Dynamic JSON object. Article arrays are wrapped in articles; other arrays in items. Explicit empty lists are valid no-match results. Empty content returns HTTP 422, invalid extraction responses HTTP 502, and unavailable service HTTP 503; these failures use business code 500. |
API Response Status Codes
| Status Code | Explanation of Status Code | Remarks |
|---|---|---|
| 200 | Request processed successfully. | Some endpoints expose a separate application-level status field in the response body, such as `dataStatus.statusCode`. |
| 400 | Invalid request parameters or request format. | Check required fields, data types, and request body format. |
| 401 | Missing or unknown application key. | Provide a valid `appkey` with the request. |
| 403 | The application key is recognized but access is not allowed. | The key may be expired, inactive, or not permitted for the requested API. |
| 429 | Request rate or trial usage limit exceeded. | Reduce concurrency or retry after the limit window resets. |
| 500 | Internal service error. | Retry later or contact support if the error persists. |
| 503 | Upstream service unavailable. | Retry later; the requested upstream dependency is temporarily unavailable. |
Remote MCP
AI client access
Use this API through GuGuData Remote MCP
Connect your AI client once, authorize in the browser, and the client can use the GuGuData API tools available to your account. You do not need to paste an appkey into the MCP client.
Endpoint
https://mcp.gugudata.io/mcpGeneric MCP configuration
{
"mcpServers": {
"gugudata": {
"url": "https://mcp.gugudata.io/mcp",
"transportType": "streamable-http"
}
}
}Code Snippets
LLM-ready Article Content Extraction API
Extract clean, LLM-ready article content and metadata
Webpage Readable Content Extraction
Extract readable articles from webpages or HTML.
Extract Links from URL
Extract hyperlinks and destinations from a webpage