A Go implementation of the Model Context Protocol (MCP) server with support for tools, resources, prompts, and other capabilities
Two tools with minimal descriptions and incomplete parameter documentation. example_tool has a trivial 23-char description ('this is an example tool') and basic schema with two required params (param1: string, param2: number) but no param descriptions visible. system_user_name has a 54-char description and empty input schema (no params). Neither tool has error handling guidance, output schema documentation, or actionable descriptions. Tool names lack clear action verbs (example_tool is generic; system_user_name is acceptable but vague on return format). No evidence of tool annotations, permission gates, or composition patterns. Baseline: 194 chars avg description, 72 chars avg param annotation, both tools fall well short.
this is an example tool
returns the username of the current logged-in user
Tool descriptions are too short and lack actionable context. 'this is an example tool' (23 chars) and 'returns the username of the current logged-in user' (54 chars) do not explain WHEN to use the tool, WHAT it returns, or any prerequisites. Baseline: 194 chars avg.
example_tool has no parameter descriptions. param1 and param2 are named generically with no guidance on valid values, format, or constraints. LLMs cannot infer meaning from bare names.
No output schema documented for either tool. LLMs need to know what fields to expect in responses to plan downstream calls and extract data correctly.
example_tool name is generic and does not start with a clear action verb. 'example_tool' conveys no intent; rename to reflect actual behavior (e.g., 'calculate_value', 'fetch_data').
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 40 | <=2025-11-25 | v2 |
No error handling guidance. Tools do not document what errors can occur, whether they are retryable, or what the LLM should do next. No recovery paths provided.