Desktop GUI application for inspecting and testing MCP servers. Allows connection to MCP servers via STDIO, SSE, or HTTP transports and provides interactive tools for listing/calling tools, resources, and prompts.
Three tools with complete input schemas and basic descriptions, but significant gaps in parameter documentation and output schema specification. Tool names follow verb_noun convention (add, greet, echo) but descriptions are minimal (10-34 chars, below the 50-200 char LLM-optimized baseline). No parameter descriptions visible in schema, LLMs cannot infer what 'a' and 'b' mean beyond type. No output schemas documented. No error handling guidance. This is a typical community server with functional but incomplete definitions.
Add two numbers together.
Echo back a message with metadata.
Greet someone by name.
Parameter descriptions missing. Schema defines 'a', 'b', 'name', 'formal', 'message' with types but no descriptions. LLMs cannot infer parameter semantics from names alone, 'a' and 'b' are ambiguous (operands? indices? flags?). Rubric baseline: 100% of A+ tools have parameter descriptions.
Tool descriptions too short (10-34 chars). Rubric baseline: 50-200 chars for LLM-optimized descriptions. Current descriptions lack context on WHEN to use each tool, WHAT it returns, and any prerequisites. 'Add two numbers together' does not explain return type, error cases, or numeric limits.
Output schemas not documented. Rubric baseline: 100% of A+ tools have documented return types. No specification of what add() returns (integer? object with result field?), what greet() returns (string? object?), or what echo() returns beyond 'metadata'.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 51 | 2026-07-28+ | v2 |
No error handling guidance. Rubric pattern:recovery-guide requires error responses to tell the LLM what to do next. No specification of how tools fail, what errors are retryable, or how to recover (e.g., if add() receives non-integers, what happens?).
Tool 'echo' has vague name and description. 'Echo' is generic; 'echo back a message with metadata' does not clarify what metadata is included, when to use it vs. other tools, or what the output structure is. Rubric: ambiguous names cause LLMs to guess wrong.