WeRead MCP has 10 tools with consistent naming (verb_noun pattern: get_*, authenticate, search_*) and basic descriptions. However, parameter descriptions are minimal or absent for most tools. Input schemas are present but lack detail, most parameters have only type and description, no constraints (enums, ranges, patterns). Output schemas are inferred from Pydantic models but not explicitly documented in tool definitions. Error handling is minimal; most tools raise generic exceptions without recovery guidance. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite all being read-only operations.

• Add detailed parameter descriptions to all tools. For book_id, specify: 'The unique identifier for a book in your WeRead library. Obtain via get_books() or search_books().' For query, specify: 'Search term (title or author name). Partial matches supported.'
• Document output schemas explicitly in tool docstrings. E.g., 'Returns: {books: [{id, title, author, cover, rating}], total_count: int}' so LLMs know what fields to expect.
• Add readOnlyHint annotations to all 10 tools via FastMCP's tool decorator to signal safety and enable caching.
• Enhance error messages with recovery paths. E.g., 'Book with ID {book_id} not found. Call get_books() to list available books.' instead of generic 'Book not found'.
• Add max_highlights parameter validation and description to format_book_highlights: 'Maximum number of highlights to include (1-100, default: 10). Higher values may exceed token limits.'