WHOIS / DNS Lookup (DoH + RDAP)

Looks up DNS records and domain registration data for any domain, using two official, documented public protocols — DNS-over-HTTPS (RFC 8484) and RDAP (RFC 7480), the structured successor to legacy WHOIS text. No login, no scraping, no anti-bot evasion: these are lookup services built to be queried by machines.

What this returns

Each domain produces one combined record:

{
  "domain": "github.com",
  "dns": {
    "records": { "A": [...], "MX": [...], "NS": [...], "TXT": [...] },
    "has_a_or_aaaa": true,
    "resolved_via": "google"
  },
  "rdap": {
    "found": true,
    "registered_at": "2007-10-09T18:20:50Z",
    "expires_at": "2026-10-09T18:20:50Z",
    "nameservers": ["NS-1283.AWSDNS-32.ORG", "..."],
    "entities": [{"roles": ["registrar"], "name": "MarkMonitor Inc."}]
  }
}

Why DoH and RDAP instead of a WHOIS scrape

Legacy WHOIS is unstructured text with a different format per registry, and its port-43 protocol is not reachable over plain HTTP at all. RDAP is its replacement: same information, structured JSON, one bootstrap redirector (rdap.org) that finds the right registry for any TLD. DNS-over-HTTPS is the same idea for name resolution — plain HTTP GET, clean JSON, no resolver configuration needed.

Two providers with automatic fallover. DNS queries try Google (dns.google) first, then Cloudflare (cloudflare-dns.com) if that fails — both are documented public services with no rate-limit surprises in normal use.

A note on "not found"

rdap.org closes the connection outright for domains with no RDAP record, rather than returning a clean 404. This Actor treats that as a fast, expected outcome (one quick retry, then a clear found: false) rather than retrying it with full backoff — a domain being unregistered, or served by a registry not yet in the bootstrap list, is normal, not an error.

Personal data is redacted, by design

Since GDPR, virtually every registry redacts registrant personal information from RDAP responses. entities in the output typically holds only the registrar — this is the registry's own privacy policy, not a limitation of this Actor, and it is also why this target is compliance-safe: no personal data is being extracted from a service that intended to protect it.

Input

Field	Type	Notes
domains	array	One or more domains.
recordTypes	array	DNS types to fetch. Default: A, AAAA, MX, NS, TXT.
includeDns	boolean	Toggle DNS lookups.
includeRdap	boolean	Toggle RDAP lookups.

Known limits

• RDAP coverage is not universal. Some ccTLD registries do not yet participate in the RDAP bootstrap; those domains return rdap.found: false even when registered. DNS results are unaffected and always attempted.
• No historical data. This is a live lookup, not a WHOIS history service — each run reflects the current state only.
• No personal registrant data. By registry policy, not by omission — see above.

Local development

pip install -r requirements.txt
python test_local.py github.com apify.com --out sample_output.json
python test_local.py example.com --record-types A MX TXT

sample_output.json in this folder is real output from a live run, kept so the schema can be reviewed without running anything.