RDAP — Registration Data Access Protocol (domains, IPs, ASNs) via IANA bootstrap
RDAP server has 5 tools with basic schemas and minimal descriptions. Tool names are nouns (domain, ip, asn, entity, nameserver) rather than action verbs (get_domain, lookup_ip, etc.), violating the verb_noun convention that 90% of production tools follow. Descriptions are extremely terse (10-30 chars), below the 50-200 char baseline for LLM-optimized descriptions. No output schema documentation. No error recovery guidance. Input schemas present but parameter descriptions entirely missing.
ASN allocation record.
Domain registration record.
Entity by handle. Pass base RDAP url to skip bootstrap.
IP/netblock allocation record.
Nameserver record.
Tool names are nouns, not action verbs. Should be get_domain, lookup_ip, query_asn, etc. LLMs infer intent from verb-first names; bare nouns force guessing.
Parameter descriptions completely missing. 'name', 'address', 'number', 'handle', 'host' have no descriptions. Rubric: 100% of A+ tools have param descriptions. LLMs cannot infer that 'address' accepts CIDR notation without explicit text.
Tool descriptions are 10-30 characters, far below 50-200 char baseline. 'Domain registration record.' and 'IP/netblock allocation record.' lack context on when to use, what it returns, and prerequisites. No WHAT/WHEN/RETURN structure.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 36 | 2026-07-28+ | v2 |
No output schema documented. LLMs cannot plan downstream calls or extract fields from RDAP JSON responses. Responses are raw API output with no field documentation.
Error messages are generic ('RDAP: not found', 'RDAP: no DNS bootstrap'). No recovery guidance. Rubric: errors must tell LLM what to do next. Should suggest 'Try a different TLD' or 'Check if domain is registered'.