MCP server implementation in Deno demonstrating clean architecture with multiple tool providers
Three tools are defined with clear naming (add, echo, hello_world), complete input schemas, and reasonable descriptions. Tool names follow verb_noun pattern adequately. However, descriptions are minimal (10-50 chars, below the 50-200 char ideal for LLM optimization), and output schemas are entirely undocumented. No error handling guidance is provided to help LLMs recover from failures. Parameter descriptions are present but sparse. The 'hello_world' tool has an optional parameter that could benefit from clearer default behavior documentation.
Add two numbers and return the result.
Echo the provided text back to the caller.
Greet the user with an optional name.
Tool descriptions are too brief (10-50 chars) and lack LLM-optimization guidance. Baselines show A+ tools average 194 chars with context for tool selection timing.
No output schemas documented. LLMs cannot plan downstream tool calls or extract structured data without knowing response structure.
No error handling guidance. Tool error responses are generic ('Unknown tool', 'Invalid argument') without actionable recovery steps for LLMs.
hello_world has an optional 'name' parameter but description does not clearly document the default value ('World') or when to use the parameter.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | F | 10 | 2026-07-28+ | v2 |
| 2026-03-09 | D | 53 | - | v1 |
Tool naming could be more specific. 'add' is acceptable but 'add_numbers' would be clearer. 'hello_world' conflates greeting and greeting targets, consider 'greet_user' to match the action.