Skip to directory

Simplescraper MCP

Simplescraper

Streamable HTTP (remote MCP; transport named in endpoint integration docs)13 documented tools16

About Simplescraper MCP

A hosted MCP service for using a Simplescraper account to discover website URLs, extract page data or Markdown, manage scraping recipes, run scrapes, retrieve results, and capture screenshots.

What you can use it for

  • Extract named product, job, or other page fields with AI.
  • Read an individual page as Markdown.
  • Create, inspect, update, and run saved scrape recipes.
  • Review previous scrape results and run history.
  • Maintain batch URL lists and capture webpage screenshots.
Connection and setup
Hosting
remote
Transport
Streamable HTTP (remote MCP; transport named in endpoint integration docs)
Endpoint
https://mcp.simplescraper.io/mcp
Authentication
OAuth sign-in through the client browser using a Simplescraper account. Current server guide says do not add a manual Authorization header or API key to the hosted endpoint configuration.

Requirements

  • A Simplescraper account; sign in through the client OAuth flow.
  • A client that supports remote MCP and the server OAuth flow.
  • Claude Desktop Integrations availability is documented as Max-plan-only at the time of the source page; availability may change.

Setup steps

  1. Add the remote MCP URL https://mcp.simplescraper.io/mcp in a client that supports remote MCP.
  2. Complete the browser sign-in/authorization flow with the Simplescraper account. Do not paste an API key or manually add an Authorization header for this hosted OAuth setup.
  3. After connection, use extract_urls for free sitemap URL discovery or request a documented recipe, extraction, Markdown, results, or screenshot operation. Paid/credit-consuming tools may consume account credits.
  4. Confirm the client reports the server connected and exposes its tools; no connection test was performed in this research.
Tools 13

13 exact tool names in the current publisher reference; schema arguments are not fully published. Hosted access uses OAuth and account credit rules.

  • run_recipe
    01
  • extract_urls
    02
  • extract_markdown
    03
  • smart_extract
    04
  • get_latest_results
    05
  • get_results
    06
  • get_results_history
    07
  • list_recipes
    08
  • get_recipe
    09
  • create_recipe
    10
  • update_recipe
    11
  • update_batch_urls
    12
  • screenshot
    13
Limitations and notes
  • Most operations consume account credits; extract_urls is explicitly described as free. Screenshot is documented as one credit per request.
  • The full remote MCP setup is OAuth/account based. An older official 2025 blog describes a local npm/API-key setup; current MCP documentation instead documents the hosted URL and OAuth flow. This legacy-vs-current setup difference is recorded and the current guide is used.
  • Claude Desktop Integrations availability is stated as Max-plan-only in the guide and could change.
  • The docs describe recipe creation and updates as capabilities, but do not expose formal input/output JSON schemas on the inspected pages.
  • No repository or package metadata for a separate public server implementation was established; the documented product is the publisher-hosted remote service.
  • No connection test was performed.
Original sources
  • official documentation

    Current Simplescraper first-party docs enumerate 13 exact hosted tools, matching the stored catalog. The endpoint is Streamable HTTP with browser OAuth; several scrape operations can consume account credits. No API key or manual Authorization header is used for the hosted OAuth connection.

  • official tool reference

    Current Simplescraper first-party docs enumerate 13 exact hosted tools, matching the stored catalog. The endpoint is Streamable HTTP with browser OAuth; several scrape operations can consume account credits. No API key or manual Authorization header is used for the hosted OAuth connection.

Documentation checked 2026-10-28T13:26:33Z. Connection has not been runtime tested.

Similar tools