Structured Web Data Extraction · MrScraper

Provide one page URL to request structured information from its retrieved content. An optional prompt describes the details or arrangement you want. Use the actual returned values to build a readable record, keeping source context and missing information visible.

Your AI

Use with your AI

For tasks with several steps, continue in the AI you already use.

CodexClaude CodeCursorWindsurfCline

View instructions

Reuse my existing Toolmu connection; if setup is needed, follow https://toolmu.com/SKILL.md. Use get_tool for extract-structured-data-mrscraper (Tool ID 5533895f-8b47-4c47-a9d7-61aa6d4a444f) and check current availability, inputs and cost against my authorized budget. Ask for the actual page URL only if it is missing. Prepare {"body":{"url":"<my actual page URL>"}}. Add an optional prompt to describe the information or arrangement I want, and choose rendering, waiting, resource or routing settings only when useful. State important option values explicitly. The prompt only parses already fetched content and cannot steer navigation or invent missing facts. Run only within my authorized budget. Present the actual returned information, preserve source values and context, and distinguish your interpretation or formatting from the source. Report the actual outcome and confirmed cost, or say when they remain unconfirmed.

Task:
Please organize the information available on this page: [paste the page link]. If there is something specific you want to know, add it in everyday words: [details to find]. Mark anything the page does not show.

Paste into your connected AI's conversation.

Connection guide

Up to 120 credits / call

The final charge is based on confirmed usage. Unused reserved credits return to your available balance.

MrScraper

Structured Web Data Extraction · MrScraper

You receive

Structured information parsed from the content retrieved at your page URL. You can request particular information or a desired arrangement with a prompt; the available fields and values depend on the page. Your Agent can present what was actually returned as a readable record, keeping supplied units and context intact.

Start with your task

Please organize the information available on this page: [paste the page link]. If there is something specific you want to know, add it in everyday words: [details to find]. Mark anything the page does not show.

What to know

A prompt cannot click through a site, control the scraper or create information that was not fetched. Rendering and waits do not guarantee access or complete content. Requested fields may be unavailable, and a desired structure does not verify the values inside it. One page request does not establish full-site coverage.

Inputs & outputs

Inspect the current inputs, availability, and maximum charge before starting.

Input schema

Input schema
Field / Path	Type	Requirement	Meaning
Extraction request`body`	`object`	Required	Required object for the page URL and useful optional parsing or loading settings. Only defined properties are accepted; url is the only required child. Required fields `url`
Page to extract`body.url`	`string`	Required	Required full URI for the single target page, including http:// or https:// as described by the supplier. No default URL. URI validity alone does not establish access, suitable content or successful parsing. Format `uri`
Real-device routing`body.super`	`boolean`	Optional	Optional boolean requesting real-device routing for stricter access conditions. Supplier-described default: false. More upstream resources and tokens are used; it is not a guarantee of access or extraction quality.
Information and form to parse`body.prompt`	`string`	Optional	Optional nonempty string describing desired information, field names or arrangement. Parses already scraped content only; it cannot control the scraper or create missing facts. No default prompt is defined in the current contract. Minimum length `1`
Page access country`body.geoCode`	`string`	Optional	Optional string of exactly two characters selecting the page-load country. Supplier-described default: us. The requested country does not verify the returned page language, currency or local context. Minimum length `2` Maximum length `2`
Page-load wait limit`body.timeout`	`integer`	Optional	Optional integer from 1 to 300 seconds limiting the wait for page loading. Supplier-described default: 300. It does not guarantee later dynamic updates or complete extraction. Minimum `1` Maximum `300`
Navigation completion event`body.waitUntil`	`string`	Optional	Optional string used with browser rendering: domcontentloaded, load, networkidle0 or networkidle2. Supplier-described default: domcontentloaded; networkidle0 is described as most thorough but slowest. The event does not verify all desired values are loaded. Options `domcontentloaded, load, networkidle0, networkidle2`
Proxy exit country`body.proxyCountry`	`string`	Optional	Optional string of exactly two characters for the proxy exit country. Supplier-described default: us. This is separate from geoCode and does not confirm localized content or successful access. Minimum length `2` Maximum length `2`
Skip images, fonts and styles`body.blockResources`	`boolean`	Optional	Optional boolean for skipping images, fonts and stylesheets during loading. Supplier-described default: false. It can reduce bandwidth use but changes page resources; consider whether needed content relies on them.
Element to wait for`body.waitForSelector`	`string`	Optional	Optional nonempty CSS-selector string. Waits for that selector to appear in the DOM before extraction, for content arriving after initial loading. Its appearance does not prove final content or a requested value is present. Minimum length `1`
Browser rendering`body.browserRendering`	`boolean`	Optional	Optional boolean to load in Chromium and render JavaScript-dependent content. Supplier-described default: false. Use when the needed information depends on rendering; it does not guarantee interaction, access or complete content.

View raw JSON

{
  "type": "object",
  "required": [
    "body"
  ],
  "properties": {
    "body": {
      "type": "object",
      "$schema": "https://json-schema.org/draft/2020-12/schema",
      "required": [
        "url"
      ],
      "~standard": {
        "vendor": "zod",
        "version": 1,
        "jsonSchema": {}
      },
      "properties": {
        "url": {
          "type": "string",
          "format": "uri",
          "description": "Full URL to scrape, including the http:// or https:// scheme."
        },
        "super": {
          "type": "boolean",
          "description": "Route through real devices to bypass hard anti-bot protection. Uses more upstream resources, so runs consume more tokens. Default false."
        },
        "prompt": {
          "type": "string",
          "minLength": 1,
          "description": "Natural-language description of the data to extract (field names, shape). Parses the scraped content only — it cannot steer the scraper itself."
        },
        "geoCode": {
          "type": "string",
          "maxLength": 2,
          "minLength": 2,
          "description": "Two-letter country code to load the page from (default 'us')."
        },
        "timeout": {
          "type": "integer",
          "maximum": 300,
          "minimum": 1,
          "description": "Maximum seconds to wait for the page to fully load (default 300)."
        },
        "waitUntil": {
          "enum": [
            "domcontentloaded",
            "load",
            "networkidle0",
            "networkidle2"
          ],
          "type": "string",
          "description": "When the browser considers navigation complete (browser rendering only). 'networkidle0' is the most thorough but slowest. Default 'domcontentloaded'."
        },
        "proxyCountry": {
          "type": "string",
          "maxLength": 2,
          "minLength": 2,
          "description": "Two-letter country code of the proxy exit (default 'us')."
        },
        "blockResources": {
          "type": "boolean",
          "description": "Skip images, fonts, and stylesheets to speed up scraping and cut bandwidth-driven token usage. Default false."
        },
        "waitForSelector": {
          "type": "string",
          "minLength": 1,
          "description": "Wait until this CSS selector appears in the DOM before extracting — for content that loads after the initial page."
        },
        "browserRendering": {
          "type": "boolean",
          "description": "Render the page in a real Chromium browser. Required for JavaScript-heavy sites and SPAs. Default false."
        }
      },
      "additionalProperties": false
    }
  },
  "additionalProperties": false
}

Result schema

Output fields have not been documented yet.

Additional notes

The task takes one full page URL and requests structured extraction from the content retrieved there. Use it when you already know the target page and want its information organized for review. The available information varies with the page and source access; a single URL request is not a site crawl.

prompt is optional and describes the desired information, field names or form. It applies only to parsing already scraped content. It cannot instruct browser actions, follow a sequence through the site or provide a fact absent from the fetched material. A requested structure is guidance, not proof that each value exists or is correct.

Browser rendering loads JavaScript-dependent content in Chromium. waitUntil selects a navigation completion event, while waitForSelector waits for a particular element to appear. Neither establishes that every later update is complete. timeout limits page-loading wait in seconds; it is not a measure of information coverage.

Resource blocking skips images, fonts and stylesheets, which can affect what loads. geoCode and proxyCountry request access and proxy-exit countries; super requests real-device routing with greater resource use. These choices do not establish a particular returned language, local offer or successful access.

Inspect the actual field values before interpreting or transforming them. Retain any units, currency labels, dates and source context provided, and keep missing values distinct from zero or false. Agent formatting or normalization is an additional step and should be explained; it does not add evidence to the extraction.

MrScraper single-page JSON extraction guidance

Questions about this tool.

Inputs, results and details to know before running it.

Get connected

No. A prompt describes what to parse and how to arrange it, while available values depend on the fetched page. Review the actual keys, values and their context before using them. A structured form does not by itself verify completeness, accuracy or consistent types across pages.

Structured Web Data Extraction · MrScraper

Structured Web Data Extraction · MrScraper

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose the page and details

Ask for information the page may show

Check the details on the original page

Questions about this tool.

You receive

Start with your task

What to know

Inputs & outputs

Input schema

Result schema

Choose the page and details

Ask for information the page may show

Check the details on the original page

Questions about this tool.

Does a requested structure guarantee every field and value?

Can a prompt click through the site or fetch linked pages?

Is the page complete once a selector or navigation event is reached?

Should the Agent automatically normalize currencies, units or dates?

Can missing information be filled with zero, false or an estimate?