Zoty demonstrates solid tool design with clear verb-noun naming (search_library, list_collections, get_item) and comprehensive parameter schemas. All 8 tools have descriptions and typed input schemas. However, output schemas are not explicitly documented in the source, parameter descriptions lack constraint details (ranges, formats, enums), and error handling guidance is minimal. Tool composition is strong, each tool has a single responsibility and output fields chain well (e.g., search_library returns 'key' which downstream tools accept). Descriptions average ~120 chars, meeting the 10-1024 baseline, but lack actionable guidance on when to use each tool vs. alternatives.

• Document output schemas for all 8 tools. For each, specify the response structure, field types, and nested objects. Example: search_library returns {items: [{key: string, title: string, abstract: string, attachment_count: int, attachments?: [...]}], total: int, returned_count: int}.
• Add enum constraints to parameters. For 'item_type', list canonical Zotero types (e.g., 'book', 'journalArticle', 'conferencePaper'). For 'style', reference the Zotero Style Repository or provide a discovery tool (list_citation_styles). For 'limit', formalize as minimum=0, maximum=25.
• Enhance error handling descriptions. Add guidance like: 'If search_library returns no results, try broader keywords or use list_collections to verify the collection exists.' For add_paper, explain what happens if the arXiv ID or DOI is invalid.
• Consider splitting 'get_bibtex_and_citation_for_items' into separate tools (get_bibtex, get_citation, get_bibliography) if they are often called independently, or clarify in the description that all three formats are always returned together.
• Document pagination fully. Specify whether pagination uses offset/limit or cursor-based. If cursor-based, include a 'next_cursor' field in responses. If offset-based, document the offset parameter and how to calculate it from returned_count and total.