ClawHub

Install the MrScraper skill on ClawHub to give your OpenClaw agent AI-powered, unblockable web scraping through natural language.

ClawHub Integration

ClawHub is a marketplace of skills and plugins for OpenClaw agents. Skills extend an agent's capabilities across research, integrations, and automation, and are installed from a central registry.

The MrScraper skill lets your agent run AI-powered, unblockable web scraping and data extraction using natural language, backed by the MrScraper API.

Overview

Once installed, your agent can:

Unblock pages: Retrieve HTML from sites that block ordinary requests, using a stealth browser and IP rotation.
Create AI scrapers: Turn a natural-language instruction into a working scraper.
Rerun scrapers: Apply an existing scraper configuration to new URLs, one at a time or in bulk.
Run manual scrapers: Execute browser workflows built from selectors.
Fetch results: Retrieve scrape results, paginated or by ID.

The skill makes direct HTTPS calls to the MrScraper API. There are no bundled scripts and no local installation step beyond installing the skill itself.

Prerequisites

An OpenClaw agent with the openclaw CLI available.
A MrScraper API token.

Installation

Install the skill

openclaw skills install @ai-mrscraper/mrscraper

Create a MrScraper API token

Open the MrScraper dashboard.
Click your user profile in the top-right corner.
Select API Tokens.
Click New Token.
Enter a name and an expiration date.
Click Create and copy the token.

Store the token

Expose the token to your agent as the MRSCRAPER_API_TOKEN environment variable.

export MRSCRAPER_API_TOKEN="your-token-here"

Keep your token secret

Never expose your API token in client-side code, logs, or commits. Store it in an environment variable or a server-side secret manager.

Authentication

Every request the skill makes carries these headers:

x-api-token: <MRSCRAPER_API_TOKEN>
accept: application/json
content-type: application/json

MrScraper exposes two base URLs:

API	Base URL	Purpose
Unblocker	`https://api.mrscraper.com`	Fetch HTML from blocked pages
Platform	`https://api.app.mrscraper.com`	Create, rerun, and read scrapers and results

Available Endpoints

Unblocker

Opens blocked pages using a stealth browser and IP rotation, and returns the page content.

curl 'https://api.mrscraper.com?url=https%3A%2F%2Fexample.com&timeout=120&geoCode=SG' \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>"

Parameter	Required	Default	Description
`url`	Yes	—	The target URL, URL-encoded.
`timeout`	No	`60`	Maximum seconds to wait for the page to load.
`geoCode`	No	—	ISO country code for proxy routing.
`blockResources`	No	—	Block images, CSS, and fonts for faster loading.

Create AI Scraper

Creates a scraper from a natural-language instruction and runs it.

curl -X POST "https://api.app.mrscraper.com/api/v1/scrapers-ai" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "url": "https://books.toscrape.com/catalogue/a-light-in-the-attic_1000/index.html",
    "message": "Extract title, price, stocks, and rating",
    "agent": "general"
  }'

Agent types:

Agent	Description
`general`	Standard pages such as product details, articles, and profiles. This is the default.
`listing`	Product listings, job boards, search results, and other paginated collections.
`map`	Website crawling that follows links to discover pages.

Rerun AI Scraper

Applies an existing scraper configuration to a new URL.

curl -X POST "https://api.app.mrscraper.com/api/v1/scrapers-ai-rerun" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "scraperId": "6695bf87-aaa6-46b0-b1ee-88586b222b0b",
    "url": "https://shopee.sg/"
  }'

Bulk Rerun AI Scraper

Runs one scraper configuration across multiple URLs in a single request.

curl -X POST "https://api.app.mrscraper.com/api/v1/scrapers-ai-rerun/bulk" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "scraperId": "6695bf87-aaa6-46b0-b1ee-88586b222b0b",
    "urls": [
      "https://books.toscrape.com/catalogue/a-light-in-the-attic_1000/index.html",
      "https://books.toscrape.com/catalogue/tipping-the-velvet_999/index.html"
    ]
  }'

Manual Scraper Rerun

Executes a browser workflow built from selectors, for cases where you want explicit extraction rules instead of AI.

curl -X POST "https://api.app.mrscraper.com/api/v1/scrapers-manual-rerun" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>" \
  -H "Content-Type: application/json" \
  -d '{
    "scraperId": "6695bf87-aaa6-46b0-b1ee-88586b222b0b",
    "url": "https://books.toscrape.com/",
    "workflow": [
      {
        "type": "extract",
        "data": {
          "extraction_type": "text",
          "name": "book",
          "selector": "h3 a"
        }
      }
    ]
  }'

Fetch Results

Returns paginated scrape results.

curl -X GET "https://api.app.mrscraper.com/api/v1/results?sortField=updatedAt&sortOrder=DESC&page=1&pageSize=10" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>"

Fetch Result by ID

Returns the detail of a single result.

curl -X GET "https://api.app.mrscraper.com/api/v1/results/497f6eca-6276-4993-bfeb-53cbbbba6f08" \
  -H "x-api-token: <MRSCRAPER_API_TOKEN>"

Map Agent Configuration

When you use the map agent, these parameters control how the crawl behaves:

Parameter	Description
`maxDepth`	How many levels of links to follow. A value of 1–2 is recommended.
`maxPages`	Maximum number of pages to crawl.
`limit`	Maximum number of records to extract.
`includePatterns`	Regex patterns for URLs to include. Separate multiple patterns with `\|\|`.
`excludePatterns`	Regex patterns for URLs to exclude. Separate multiple patterns with `\|\|`.

Keep crawls bounded

Set maxDepth to 1 or 2 and cap maxPages. Depth grows the crawl exponentially, and an unbounded crawl burns tokens fast.

Error Handling

The API uses standard HTTP status codes:

Status	Meaning
`400`	Invalid request payload.
`401`	Missing or invalid API token.
`404`	Resource not found.
`429`	Rate limit exceeded.
`500`	Internal error.

Retry 429 responses with exponential backoff.

Data Scope

Data is transmitted only to MrScraper servers. Responses contain the extracted page content and its metadata. Never expose your API token in logs or commits.

ClawHub

On this page