Loading ToolBench
Model Context Protocol for Realtime Register API.
Server has 7 tools with explicit schemas and descriptions. Naming follows verb_noun convention (list_domains, get_domain, register_domain, etc.). However, critical gaps exist: (1) Two tools (confirm_pending_action, cancel_pending_action) have EMPTY input schemas with no parameters documented, violating the requirement that every parameter needs description. (2) Descriptions are present but generic, most are 40-80 chars, below the 50-200 char LLM-optimized baseline. (3) Parameter descriptions in list_domains and list_contacts reference 'Invalid matcher' for the matcher field, which is incomplete and unhelpful. (4) No output schemas documented, LLMs cannot infer what fields to expect from responses. (5) Error handling is absent, no recovery guidance, no categorization of retryable vs fatal errors. (6) register_domain accepts optional authcode and skipValidation but lacks explanation of when/why to use them. Schemas are present and use Zod validation, but descriptions do not guide LLM decision-making.
Cancel a pending action
Confirm a pending action
Show contact information by handle
Show domain information by domain name available within your account.
List contacts in your account
List domains within your account
Checks if a domain name is available and registers it if available
Two tools (confirm_pending_action, cancel_pending_action) have empty input schemas with no parameters documented. Per HARD SCORING RULE, schema score MUST be 0 when no schema is visible.
No output schemas documented for any tool. LLMs cannot infer response structure, forcing them to parse unstructured JSON and guess field names for downstream tool calls.
Parameter descriptions are generic or incomplete. 'Invalid matcher' in list_domains/list_contacts filters is unhelpful. register_domain lacks explanation of when to use quote, skipValidation, or authcode.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2025-06-18+ | v2 |
No error handling or recovery guidance. Tools do not indicate which errors are retryable, user-fixable, or fatal. No actionable error messages documented.
Descriptions are too short (40-80 chars vs 50-200 char baseline). Per HARD SCORING RULE, descriptions under 20 chars score 0-20; these are marginally better but still below LLM-optimized length for clear tool selection.