Server has 5 well-named tools with clear descriptions and documented input schemas. All tools follow verb_noun naming (wow_character_audit, wow_evaluate_player, etc.) and include region/realm/name parameters with descriptions. However, output schemas are not documented, LLMs cannot predict response structure. Parameter descriptions are adequate but lack format constraints (e.g., realm slug format, region enum). No error handling guidance visible. Tool descriptions are in French, which may confuse English-speaking LLMs. Missing tool annotations (readOnlyHint, idempotentHint) despite all tools being read-only.

• Document output schemas for each tool. E.g., wow_character_audit returns {name, realm, region, mythicPlusScore, equippedILevel, raidProgression, missingEnchants[], missingGems[], bestRuns[], weakestDungeon, timedRatio}. This enables LLMs to extract and chain data.
• Translate all tool descriptions to English. Use clear, concise descriptions (50-200 chars) that explain WHAT the tool does, WHEN to use it, and WHAT it returns. E.g., 'Fetch a WoW character's Mythic+ score, gear, enchants, gems, talents, and recent runs. Identifies missing enchants/gems and weakest dungeons.'
• Add enum constraints to region parameter: {type: 'string', enum: ['eu', 'us', 'kr', 'tw'], description: 'Region code (default: eu)'}. This prevents invalid input and self-documents valid options.
• Add tool annotations to all 5 tools: @Tool(..., readOnlyHint=true, idempotentHint=true). This signals to agents that these tools are safe to retry and do not modify state.
• Implement error handling with recovery guidance. E.g., if character not found, return: 'Character "name" not found on realm "realm" (region: eu). Verify spelling and realm slug format (lowercase, dashes for spaces). Try wow_search_characters() to find the correct realm.'