MCP server for DNS lookups, reverse DNS, WHOIS, and domain checks — zero auth, zero config
The server defines 5 DNS/WHOIS tools with clear, specific action verbs (dns_lookup, reverse_dns, resolve_all, check_nameservers, whois). All tool descriptions are present and reasonably detailed (80-95 chars), explaining what each tool does. Input schemas use Zod with proper type definitions and descriptions for all parameters. However, there are notable gaps: (1) No output schemas are documented, callers cannot infer the structure of returned data; (2) Error handling is minimal, the code catches errors generically but provides no recovery guidance to LLMs; (3) No input validation or constraint messaging, e.g., 'domain' parameter has no documented format constraints, length limits, or validation rules; (4) Tool descriptions lack strategic context, they don't indicate when to use resolve_all vs dns_lookup, or how tools chain together. Parameter descriptions are adequate but could be more prescriptive (e.g., 'IP address (IPv4 or IPv6 format, e.g. 8.8.8.8 or 2001:4860:4860::8888)' instead of just 'IP address (e.g. 8.8.8.8)').
Get the nameserver (NS) records for a domain.
Look up DNS records for a domain. Returns A, AAAA, MX, TXT, NS, CNAME, SOA, SRV, or PTR records.
Resolve all common DNS record types (A, AAAA, MX, TXT, NS, CNAME) for a domain in one call.
Perform a reverse DNS (PTR) lookup on an IP address.
Query WHOIS data for a domain. Uses IANA and follows referrals to registrar WHOIS servers.
No output schemas documented. Tools return text blobs via { content: [{ type: 'text', text: result }] } without specifying the structure, format, or fields of 'result'. LLMs cannot reliably parse or extract specific data (e.g., extracting the MX priority from dns_lookup output requires string parsing).
Minimal error handling with no recovery guidance. Functions use try-catch but do not differentiate between retryable errors (network timeout), user-fixable errors (invalid domain format), and fatal errors (DNS server unreachable). Errors are not contextualized for LLM reasoning.
Parameter descriptions lack constraints and validation rules. 'domain' is described only as 'Domain name to look up (e.g. example.com)', no mention of valid format, length limits, allowed characters, or internationalized domain support. Example values are embedded in descriptions, which LLMs may reuse literally.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 57 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 44 | - | v1 |
Tool descriptions do not provide strategic guidance on when to use which tool. 'dns_lookup' and 'resolve_all' both resolve domains but descriptions don't explain: dns_lookup is for a single record type, resolve_all for all common types in one call. This risks inefficient multi-tool chains.
No timeout or rate-limit handling visible in whois implementation. The socket-based WHOIS query has a 10-second timeout, but this is not documented in the tool description. Long-running queries could block agents. No guidance provided for retries or rate limiting.