generated: '2026-08-09'
method: searched
status: published
source: https://webscraping.ai/docs
servers:
- name: webscraping-ai-hosted
  kind: remote
  transport: http
  protocol: streamable-http
  url: https://mcp.webscraping.ai/mcp
  auth: oauth2
  auth_note: 'OAuth 2.1 login with the WebScraping.AI account — no API key. Anonymous POST of tools/list
    returns 401 with `WWW-Authenticate: Bearer resource_metadata="https://mcp.webscraping.ai/.well-known/oauth-protected-resource/mcp"`.'
  oauth:
    authorization_server: https://webscraping.ai
    protected_resource: https://mcp.webscraping.ai/mcp
    metadata:
    - well-known/webscraping-ai-oauth-authorization-server.json
    - well-known/webscraping-ai-oauth-protected-resource.json
  install: claude mcp add --transport http webscraping-ai https://mcp.webscraping.ai/mcp
  docs: https://webscraping.ai/docs
- name: webscraping-ai-mcp
  kind: local
  transport: stdio
  package: webscraping-ai-mcp
  registry: npm
  version: 1.0.7
  mcp_name: io.github.webscraping-ai/webscraping-ai
  install: npx -y webscraping-ai-mcp
  auth: api-key
  auth_note: WEBSCRAPING_AI_API_KEY environment variable.
  repository: https://github.com/webscraping-ai/webscraping-ai-mcp-server
  registry_manifest: https://github.com/webscraping-ai/webscraping-ai-mcp-server/blob/master/server.json
  smithery: https://github.com/webscraping-ai/webscraping-ai-mcp-server/blob/master/smithery.yaml
  license: MIT
tools_source: Tool names, descriptions and input schemas below are read verbatim from the provider's own
  open-source server registration (https://github.com/webscraping-ai/webscraping-ai-mcp-server/blob/master/src/index.js).
  The docs state the hosted server "exposes the same 7 tools". Live `tools/list` against the hosted server
  is OAuth-gated (401), so the hosted server's schemas were not introspected.
tools:
- name: webscraping_ai_question
  description: Get an answer to a question about a given web page (LLM-powered).
  inputs:
  - url
  - question
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getQuestion
- name: webscraping_ai_fields
  description: Extract structured data fields from a web page using natural-language field instructions.
  inputs:
  - url
  - fields
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getFields
- name: webscraping_ai_html
  description: Get the full HTML of a page by URL, with proxies and Chromium JS rendering.
  inputs:
  - url
  - return_script_result
  - format
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getHTML
- name: webscraping_ai_text
  description: Get the visible text of a page by URL.
  inputs:
  - url
  - text_format
  - return_links
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getText
- name: webscraping_ai_selected
  description: Get the HTML of a selected page area by URL and CSS selector.
  inputs:
  - url
  - selector
  - format
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getSelected
- name: webscraping_ai_selected_multiple
  description: Get the HTML of multiple page areas by URL and an array of CSS selectors.
  inputs:
  - url
  - selectors
  - <common options>
  source_operation: openapi/webscraping-ai-openapi.yml#getSelectedMultiple
- name: webscraping_ai_account
  description: Information about the account API credits quota.
  inputs: []
  source_operation: openapi/webscraping-ai-openapi.yml#account
common_options:
  description: Shared across every tool except webscraping_ai_account.
  fields:
  - timeout
  - js
  - js_timeout
  - wait_for
  - proxy
  - country
  - custom_proxy
  - device
  - error_on_404
  - error_on_redirect
  - js_script
crosswalk: mcp/webscraping-ai-tool-crosswalk.yml
x-evidence:
  fetched: '2026-08-09'
  probes:
  - url: https://mcp.webscraping.ai/mcp
    method: POST
    body: tools/list
    http_status: 401
  - url: https://mcp.webscraping.ai/.well-known/oauth-protected-resource
    http_status: 200
  - url: https://raw.githubusercontent.com/webscraping-ai/webscraping-ai-mcp-server/master/src/index.js
    http_status: 200
  - url: https://registry.npmjs.org/webscraping-ai-mcp
    http_status: 200
deployment:
  mode: remote
  endpoint: https://mcp.webscraping.ai/mcp
  verified: probed
  probe: gated
  checked: '2026-08-12'
  source: catalog MCP census