niwashi-mcp demonstrates solid definition quality with all 5 tools having clear names, descriptions, and input schemas. Tool names follow verb_noun convention (search_classes, get_schema, query, get_by_dn, count). Descriptions are substantive (100-200 chars typical), explaining WHAT each tool does and WHEN to use it. Input schemas are properly typed with parameter descriptions. However, output schemas are not documented, LLMs cannot predict response structure for planning. Error handling guidance is absent. Parameter constraints (e.g., limit bounds, filter_expr syntax) lack formal specification. The query() tool is complex with 10 parameters; some interdependencies (filters vs filter_expr, include_children vs rsp_subtree_include) are underdocumented.

• Document output schemas for all tools. Specify fields, types, and pagination structure. Example: 'Returns {results: [{dn, attributes}], total_count, next_page_token}'.
• Add error handling guidance to each tool description. Example: 'If class_name not found, returns error code 404 with suggestion to call search_classes(). Retryable: no.'
• For query(), explicitly document parameter interdependencies in descriptions. Example: 'filters and filter_expr are mutually exclusive; use filters for simple equality, filter_expr for complex boolean logic.'
• Add numeric bounds to limit, page, list_limit parameters. Example: 'limit: integer, 1 - 1000 (default 20). Larger values may timeout.'
• Provide grammar or examples for filter_expr and order_by. Example: 'filter_expr uses APIC syntax: wcard(attr,pattern), ne(attr,value), gt(attr,num). See APIC docs for full grammar.'

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry