Mojeek Search Scraper

Collect normalized Mojeek web results from one search or a fair multi-query batch. The Actor enters the requested query through Mojeek's search interface, follows validated native result pages, and stores only verified result records.

Features

• Single, batch, and single-item modes
• Global and per-query record limits
• Native offset pagination with language, region, and safe-search filters preserved
• Stable URL-derived IDs and cross-query deduplication
• Consistent browser fingerprints, cookie-persistent sessions, bounded retries, and configurable concurrency
• Optional Apify Residential or user-provided proxies
• Atomic output: block, challenge, or malformed result pages are never stored as data
• A 49-field documented schema; optional values are omitted when Mojeek does not expose them

Mojeek often rejects datacenter IPs. Apify Residential proxy is recommended for cloud runs. GOOGLE_SERP cannot be used because it only supports eligible Google Search and Google Shopping requests, not Mojeek.

Input

Provide query, queries, or both. Duplicate queries are removed case-insensitively.

{
  "queries": ["artificial intelligence", "climate science"],
  "maxItems": 20,
  "maxItemsPerQuery": 10,
  "maxPages": 2,
  "language": "en",
  "region": "GB",
  "safeSearch": 0,
  "maxConcurrency": 2,
  "proxyConfiguration": {
    "useApifyProxy": true,
    "apifyProxyGroups": ["RESIDENTIAL"],
    "apifyProxyCountry": "GB"
  }
}

Set maxItems and maxItemsPerQuery to 1 for single-item mode. maxConcurrency is bounded to 1–4, retries to 0–3, queries to 25, and pagination to 50 pages per query.

Output

Every record includes a stable ID, canonical result URL, title, position metadata, query context, requested filters, extraction method, and ISO-8601 scrape time. When present on the page, records also contain snippets, sanitized highlight HTML, displayed dates, images, tags, site-cluster metadata, a result count, and a mapped knowledge panel.

Empty optional values, fabricated favicons, internal browser objects, challenge pages, cookies, authorization data, and proxy credentials are not emitted. Results are available from the default Apify dataset in JSON, CSV, Excel, and API formats.

Reliability behavior

• HTTP status and HTML content type are validated before parsing.
• Explicit 403, 429, CAPTCHA, and verification pages fail closed and mark the session bad.
• Pagination links must remain on Mojeek, preserve the query, and advance by a bounded offset.
• Temporary transport failures use bounded retries; explicit access controls are not retried or bypassed.
• A legitimate no-results page completes with an empty dataset.