# ScraperWatch

> Find web scrapers that work; compare scraping APIs, proxy prices and RAG extractors.

- name: com.scraperwatch/scraperwatch
- version: 1.2.0
- connection class: R0 (autonomous)
- trust: 83/100 flags: no-repository
- quality: 83/100 (new)
- owner: not claimed
- descriptor: https://api.protogrid.dev/v1/servers/com.scraperwatch%2Fscraperwatch
- tools: https://api.protogrid.dev/v1/servers/com.scraperwatch%2Fscraperwatch/tools

**An agent can connect right now, no human step.** Remote endpoint, no authentication, reachable on the last probe.

## Connection (mcpServers)

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


## Trust

- hygiene: 80
- liveness: 100
- freshness: 100
- provenance: 55
- drivers: +dns-namespace +website +reachable +uptime-100% +updated-1d-ago -no-repository
- Derived from observable signals (official registry feed and our own credential-free probes); no code audit performed.

## Quality 83/100 (new)

### protocol: 88 (weight 25)

- pass `protocol.modern`: supports 2026-07-28 (server/discover)
- pass `protocol.stateless`: answers without a session
- pass `protocol.transport`: streamable HTTP
- warn `protocol.list_ttl`: tools/list declares ttlMs 0, so clients re-fetch it on every use

### authorization: n/a (weight 15)

- n/a `auth.prm`, `auth.as_metadata`, `auth.cimd`: no remote uses OAuth
- n/a `auth.secret_in_url`: no templated URL

### tool hygiene: 79 (weight 30)

- pass `tools.descriptions`: every tool has a description
- pass `tools.description_length`: descriptions are concise
- pass `tools.schemas`: every tool has an object input schema
- warn `tools.annotations`: only 3 of 7 tools declare readOnlyHint or destructiveHint
- fail `tools.directory_hints`: no tool declares all four of readOnlyHint, destructiveHint, idempotentHint and openWorldHint; missing: compare_scrapers (idempotentHint); compare_proxy_providers (idempotentHint); best_scraper (readOnlyHint, destructiveHint, idempotentHint, openWorldHint); …
- pass `tools.token_cost`: about 885 tokens to load every tool
- pass `tools.api_dump`: tools are not a one-to-one API dump

### stability: n/a (weight 15)

- n/a `stability.changes`, `stability.rug_pull`: tool history recorded for 1 of the 7 days needed

### dependencies: n/a (weight 15)

- n/a `deps.known_vulns`, `deps.mcp_sdk_version`, `deps.resolvable`: no npm or PyPI package

- tools: 7, about 885 tokens to load them all
- tool set hash: 542e2147c931b418b43d93ab5bc3de33
- tools last changed: 2026-10-20T07:59:46.056Z
- badge: [![protogrid quality](https://protogrid.dev/badge/com.scraperwatch/scraperwatch.svg)](https://protogrid.dev/servers/com.scraperwatch/scraperwatch)
- Checks run on what our credential-free, read-only probes observe; tools are never called and no code audit is performed.

### Tool changes

- 2026-10-20 `compare_rag_extractors` changed (description)
  - before: Read saved Firecrawl, Jina Reader and DIY RAG extraction measurements, source dates, limitations and disclosed buyer links. No provider requests; unknown measurements are null.
  - after: Read saved Firecrawl, Jina Reader and DIY RAG extraction measurements, source dates, named legacy literal table rates and hash-verified escaped-underscore audit (otherwise unavailable/stale), limitations and disclosed buyer links. The legacy table gap is not demonstrated structural loss. No provider requests; unknown measurements are null.
- 2026-10-20 `compare_scraping_apis` changed (description) **meaning changed**
  - before: Independent benchmark of web scraping APIs (ScrapingBee, ScraperAPI, Scrapingdog, ...): success rate on test pages incl. anti-bot challenges, latency, updated every 3 days.
  - after: Saved scraping API aggregate summary plus a historical matched ScraperAPI/ScrapingBee comparison with separate standard/premium denominators, content checks, latency population, limitations and affiliate disclosure. No vendor requests.
- 2026-10-20 `best_scraper` changed (description) **meaning changed**
  - before: One recommended scraper for a website: the most reliable tool with real usage (300+ runs, 20+ users in 30 days), plus the most used and the cheapest reliable alternative.
  - after: Recommend from saved 30-day aggregate statistics: highest recorded success rate among tools with 300+ runs and 20+ users, plus most used and cheapest qualifying alternatives. Evidence includes the snapshot date and age; this is not a live execution test or assurance of current operation.
- 2026-10-20 `compare_scrapers` added
  - Compare two to five exact actor IDs from the filtered public export in caller order. Saved usage, eligible success statistics and compatible standard result-event rates only; unknown/excluded IDs and duplicates are explicit. Different targets are different workloads, with no winner. Includes snapshot age, billing caveats, affiliate disclosure and existing target pages. No actor runs or live requests.
- 2026-10-20 `find_scrapers` changed (description, inputSchema) **meaning changed**
  - before: Rank ready-made scrapers for a website by real 30-day usage, with success rate and price. site: website name or domain, e.g. 'zillow', 'amazon.com', 'google maps', 'linkedin jobs'.
  - after: Rank ready-made scrapers by saved 30-day aggregate usage, success rate and recorded price. Evidence includes the snapshot date and age, not a live execution test. min_success_rate must be a finite number from 0 to 1 inclusive; zero includes unknown rates, while a positive minimum requires a known rate at or above the threshold. site: website name or domain, e.g. 'zillow', 'amazon.com', 'google maps', 'linkedin jobs'.
- 2026-10-19 `compare_proxy_providers` added
  - Compare free cached proxy pricing offers by product type and optional positive monthly GB. Full captured package bills only; per-IP offers are separate, unknown costs stay unknown. Includes observed dates, sources, cautions and disclosed Webshare affiliate link. No live lookup.
- 2026-10-19 `compare_rag_extractors` added
  - Read saved Firecrawl, Jina Reader and DIY RAG extraction measurements, source dates, limitations and disclosed buyer links. No provider requests; unknown measurements are null.
- 2026-10-19 `compare_scraping_apis` added
  - Independent benchmark of web scraping APIs (ScrapingBee, ScraperAPI, Scrapingdog, ...): success rate on test pages incl. anti-bot challenges, latency, updated every 3 days.

Full history: https://api.protogrid.dev/v1/servers/com.scraperwatch%2Fscraperwatch/changes

## Remotes

- https://scraperwatch.com/mcp (streamable-http, auth none, reachable true, uptime30d 1)

## Packages

_none_

## Tools (7)

- `best_scraper`: Recommend from saved 30-day aggregate statistics: highest recorded success rate among tools with 300+ runs and 20+ users, plus most used and cheapest qualifying alternatives. Evidence includes the snapshot date and age; this is not a live execution test or assurance of current operation.
- `compare_proxy_providers`: Compare free cached proxy pricing offers by product type and optional positive monthly GB. Full captured package bills only; per-IP offers are separate, unknown costs stay unknown. Includes observed dates, sources, cautions and disclosed Webshare affiliate link. No live lookup.
- `compare_rag_extractors`: Read saved Firecrawl, Jina Reader and DIY RAG extraction measurements, source dates, named legacy literal table rates and hash-verified escaped-underscore audit (otherwise unavailable/stale), limitations and disclosed buyer links. The legacy table gap is not demonstrated structural loss. No provider requests; unknown measurements are null.
- `compare_scrapers`: Compare two to five exact actor IDs from the filtered public export in caller order. Saved usage, eligible success statistics and compatible standard result-event rates only; unknown/excluded IDs and duplicates are explicit. Different targets are different workloads, with no winner. Includes snapshot age, billing caveats, affiliate disclosure and existing target pages. No actor runs or live requests.
- `compare_scraping_apis`: Saved scraping API aggregate summary plus a historical matched ScraperAPI/ScrapingBee comparison with separate standard/premium denominators, content checks, latency population, limitations and affiliate disclosure. No vendor requests.
- `find_scrapers`: Rank ready-made scrapers by saved 30-day aggregate usage, success rate and recorded price. Evidence includes the snapshot date and age, not a live execution test. min_success_rate must be a finite number from 0 to 1 inclusive; zero includes unknown rates, while a positive minimum requires a known rate at or above the threshold. site: website name or domain, e.g. 'zillow', 'amazon.com', 'google maps', 'linkedin jobs'.
- `list_websites`: List the websites ScraperWatch tracks, with the number of scrapers and total 30-day users for each.