Server has 6 tools with basic descriptions and parameter schemas, but significant gaps limit production readiness. All tools are READ_ONLY and well-intentioned, but descriptions lack LLM-optimization guidance (e.g., 'when to use this vs similar tools'), parameter descriptions are minimal, and output schemas are entirely undocumented. No error handling guidance, no recovery paths, and no tool annotations. Naming is verb-forward (search_, check_, get_, browse_, list_, export_) which is good, but parameter descriptions are sparse and don't explain constraints or formats.

• Add explicit output schemas to all tools. Document the JSON structure returned (e.g., search_libraries returns {results: [{title, author, isbn, publisher}], total_count, next_cursor}).
• Convert parameter constraints to enums and numeric bounds. Change 'query_type' description to enum: ['title', 'author', 'isbn', 'subject', 'keyword']. Add min/max to max_results (1-100).
• Add error handling guidance to each tool description. E.g., 'If library is unreachable, returns {error: "Library unavailable", retry_after_seconds: 30}. Retry after delay.'
• Expand tool descriptions with 'when to use' context. E.g., search_libraries: 'Use for keyword/title/author searches across multiple libraries. For browsing by subject category, use browse_subjects instead.'
• Add readOnlyHint: true to all tool schemas to signal idempotency and safety to LLMs.

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

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