Server has 9 tools with explicit schemas and descriptions in Chinese. Naming follows verb_noun pattern (send_email, read_emails, search_emails, delete_email, reply_email, list_folders, find_email_folder, get_correspondence, web_login). However, descriptions are minimal (10-80 chars, well below the 194-char baseline), parameter descriptions lack depth, output schemas are undocumented, and error handling is absent. No tool annotations (readOnlyHint/destructiveHint). Schemas are present but sparse, many parameters lack type constraints, enums, or validation guidance. The web_login tool exposes authentication flow as a tool parameter (smsCode), which is a security concern. Overall structure is functional but falls short of production quality.

• Document output schema for every tool. E.g., read_emails should return {emails: [{id, from, subject, date, preview, unread}], total_count, has_more}. This enables LLMs to chain tools correctly.
• Expand tool descriptions to 100 - 200 chars. Add context: 'Read emails from a folder. Returns up to N emails sorted by date. Use search_emails for keyword matching.' This helps LLMs choose the right tool.
• Add enum constraints for folder parameter. E.g., folder: {type: 'string', enum: ['INBOX', 'SENT', 'DRAFTS', 'TRASH', 'CUSTOM'], description: 'Folder name. Call list_folders to discover custom folders.'}
• Add tool annotations: mark read_emails, search_emails, list_folders, find_email_folder, get_correspondence with readOnlyHint=true; mark send_email, reply_email with destructiveHint=false (they are writes but not destructive); mark delete_email with destructiveHint=true.
• Refactor web_login to use server-side session tokens. Remove smsCode parameter; instead, return a session_id after initial login, then accept session_id + smsCode in a separate verify_sms_code tool. This keeps credentials out of tool parameters.