Skip to content
ScraperWatch logo

ScraperWatch MCP server

Which MCP servers are ScraperWatch's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-28

← Back to ScraperWatch

Whose server it is
  • Registry namecom.scraperwatch/scraperwatchverified in the MCP registry

The registry name sits under scraperwatch.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-28
  • Stateanswerschecked 2026-10-28
  • Tools8
  • Server calls itselfScraperWatch
  • Server version1.3.0
  • Protocol version2025-06-18

On 2026-10-28 the server answered initialize and tools/list without a sign-in. First answer in 475 ms.

Should an agent connect itNot proven either way yet
  • Its tools do every use case we checked: compare Apify scrapers built for a chosen website; check how often a scraper's runs succeed.
  • Checked on 6 of the 7 days it takes to judge reliability.

Once connected

  • For "compare Apify scrapers built for a chosen website", the likely first call is find_scrapers.
  • For "check how often a scraper's runs succeed", the likely first call is actor_history.
  • Parameters are untyped or undescribed in 5 of 8 tools: check arguments before calling.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 6, the first on 2026-10-28
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 6 so far
  • What an agent can do with it2 of 2 use cases
  • Answeredmeasuring6 of the 7 daily checks it takes to print a share
  • Median first answer486 ms
  • Space in an agent's context≈ 1,146 tokensthe definitions of all its tools (8), which an agent carries on every turn
  • Tool definitionsB · 81 of 100
  • Typed results1 of 8 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 5 of 8
  • Not saying whether it only reads or can destroy: 4 of 8

What an agent can do with it

  • can do · compare Apify scrapers built for a chosen website
  • can do · check how often a scraper's runs succeed
  • not sure · compare proxy prices for a scraper already running

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

None of the five warning signs in these definitions.

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers8 tools · read 2026-10-28
  • find_scrapers

    Which tracked scrapers have the most recorded 30-day users for a website? Uses the filtered official Apify Store export; data_updated is the saved snapshot timestamp, with age/status in evidence. Returns usage, run completion and billing…

  • compare_scrapers

    How do two to five exact actor IDs compare? Uses the filtered official Apify Store export; data_updated and evidence describe the saved snapshot date and age. Reports usage, eligible run completion and compatible result-event rates…

  • best_scraper

    Which tracked actors lead recorded run completion, usage and qualifying price for a website? Uses the filtered official Apify Store export; data_updated and evidence give snapshot date and age. Completion candidates require 300+ runs and…

  • compare_scraping_apis

    What do saved scraping API tests report? Uses the saved API benchmark summary and frozen matched ScraperAPI/ScrapingBee report. generated_at/window_days date the aggregate only; matched evidence has its own window. Returns…

  • compare_rag_extractors

    How did Firecrawl, Jina Reader and DIY extraction score on saved documentation fixtures? Uses the saved RAG benchmark and hash-checked table audit. Per-vendor last_run dates are source observations; generation time is unknown. Returns…

  • compare_proxy_providers

    What captured proxy offers cover a product type and optional monthly GB? Uses research/e5/proxies.json; observed_on dates each captured offer, with observations outside the past 30 days omitted. Estimates full saved package bills only; IP…

  • actor_history

    How have one tracked actor's usage, run completion and primary event price changed by day? Uses compact official Apify Store history exports; dates are snapshot filenames, not verified per-actor observation times. days is a UTC calendar…

  • list_websites

    Which websites are in the filtered scraper export, and how many actors and summed 30-day users does each have? Uses saved official Apify Store statistics from latest.json. Counts reflect that snapshot, not live coverage; this response has…

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "scraperwatch": {
      "type": "http",
      "url": "https://scraperwatch.com/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http scraperwatch https://scraperwatch.com/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://scraperwatch.com/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.