A minimal MCP server with an echo tool, using Starlette, SSE, and MCP.
Single tool 'echo' has a valid input schema with proper type definition (string parameter 'message'), but the description is extremely minimal (8 words, well below the 20-char hard floor). No parameter descriptions exist, the schema lacks the 'description' field for the 'message' parameter. No output schema is documented. The tool is trivial (read-only echo) and lacks any guidance on when/why to use it or what it returns. Naming is generic ('echo') rather than action-verb specific. No error handling, no composition guidance, no security considerations.
Echo a message back to the client
Tool description is 8 characters ('Echo a message back to the client'), below 20-char hard floor. LLM cannot determine when/why to select this tool.
Input parameter 'message' has no description in schema. Rubric requires every parameter to have a non-empty description explaining what it controls. LLMs cannot infer parameter meaning from names alone.
No output schema documented. Tool returns TextContent but LLM has no formal specification of return type, fields, or structure. Rubric baseline: 100% of A+ tools have documented return types.
Tool name 'echo' is generic and does not follow verb_noun convention. Rubric baseline: 90% of A+ tools start with action verb. 'echo' lacks clarity on what action occurs.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 33 | 2026-07-28+ | v2 |
No error handling guidance. Tool does not document what happens on invalid input, empty message, or edge cases. Rubric requires error responses to tell LLM what to do next.