A multi-module Spring Boot MCP server suite providing tools for address validation, IBAN validation, and Ollama-based AI services
Two tools with minimal definitions. Both have basic input schemas with typed parameters and descriptions, but tool descriptions are extremely brief (under 50 chars). No output schemas documented. No error handling guidance. No parameter constraints (enums, ranges). Naming follows verb_noun pattern but lacks clarity on what validation is performed. No evidence of idempotency, pagination, or composition patterns. Schema quality is present but sparse.
Check if the address is a valid german address
Check if the IBAN is valid
Tool descriptions are critically brief (under 50 chars). 'Check if the address is a valid german address' lacks WHEN to use, WHAT is returned, and error conditions. LLMs cannot reliably select or reason about these tools.
No output schemas documented. Callers cannot know what fields are returned (e.g., does checkAddressValid return {valid: boolean} or {valid: boolean, normalized_address: string, error: string}?). Agents cannot plan downstream calls.
No error handling guidance. If validation fails, what does the tool return? An error code? A message? How should the agent recover? No recovery hints provided.
Input parameters lack constraints. 'plz' (postal code) has no format, length, or pattern validation documented. 'iban' has no format constraint. LLMs will pass invalid values without guidance.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 33 | 2026-07-28+ | v2 |
No tool annotations (readOnlyHint, idempotentHint). Both tools are read-only validators but this is not explicitly declared in the schema, forcing LLMs to infer safety.