This is a minimal custom MCP server example with 5 tools focused on basic arithmetic and text processing. All tools have simple, action-verb names (add, multiply, power, text_stats, transform_text) and descriptions present. However, descriptions are extremely brief (avg ~25 chars), well below the production baseline of 194 chars. Input schemas are present with proper JSON Schema structure and typed parameters, but parameter descriptions are minimal. No output schemas are documented. The tools demonstrate basic naming conventions but lack the depth of guidance needed for LLM disambiguation and error recovery. This is a demonstration/tutorial server, not production-grade.
Add two numbers together
Multiply two numbers
Raise a number to a power
Get statistics about a text string
Transform text (uppercase, lowercase, title case)
Description lengths critically short (avg 25 chars vs production baseline 194 chars). Cannot provide LLM context needed for tool selection and proper error recovery.
No output/return schemas documented for any tool. LLMs cannot plan downstream operations or validate responses.
Parameter descriptions minimal or generic. 'First number' and 'Second number' for 'a' and 'b' do not hint at accepted ranges, formats, or constraints.
No error handling guidance. Tools return bare result strings with no recovery hints or categorization (retryable vs fatal).
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 50 | <=2025-11-25 | v2 |
| 2026-03-09 | F | 35 | - | v1 |
'power' naming is ambiguous. Does it raise to a power, or return current power usage? 'raise_to_power' would be clearer.