MCP Server Toolkit — Framework, Factory and Proxy for building MCP servers
Six tools with basic descriptions and simple schemas. All tools have names starting with action verbs (echo, greet, demo_*) and descriptions present. However, descriptions are minimal (10-50 chars), lacking context on WHEN to use each tool or what distinguishes them. No parameter descriptions visible in schema definitions. No output schemas documented. No error handling guidance. Tools are READ_ONLY but lack explicit idempotentHint annotations. Schema completeness varies: echo/echo_upper/greet have single string params with minimal type info; demo_time/demo_status have empty input objects (acceptable but underdocumented); demo_summarize has a text param. Overall structure is present but shallow, typical of a demo/example server rather than production-grade tooling.
Server status as a resource.
Summarize the given text.
Return the current server time.
Return the message — for testing.
Return the message in uppercase.
Greet someone by name.
Descriptions are too brief (10-50 chars). Rubric baseline is 194 chars avg; descriptions lack WHEN to use, prerequisites, and return value guidance. LLMs cannot reliably select tools without context.
No parameter descriptions visible in schemas. Rubric requires every parameter to have a description explaining what it controls. 'message' and 'text' params lack context on format, length, or constraints.
No output schemas documented. Rubric requires documenting return types so LLMs know what fields to expect and can plan downstream calls. No guidance on what echo/greet/demo_* return.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 45 | <=2025-11-25 | v2 |
No error handling guidance. Tools lack recovery hints (e.g. 'If summarization fails, try a shorter text'). Rubric requires error responses to tell LLM what to do next.
No tool annotations (idempotentHint, readOnlyHint). All tools are READ_ONLY but lack explicit annotations. Rubric rewards tool annotations as current spec pattern.