This server has critical definition quality gaps. All 10 tools lack parameter descriptions entirely, the rubric requires descriptions for every parameter to guide LLM behavior. Tool descriptions are extremely terse (5-20 chars), well below the 10-1024 char baseline and the 50-200 char LLM-optimized target. No output schemas are documented. Parameter types are present but lack constraints (enums, ranges, formats). The 'ids' parameter appears in 6 tools but is never described, LLMs cannot infer whether to pass comma-separated strings, arrays, or individual values. No error handling guidance. No tool annotations (readOnlyHint, etc.). The server is a thin wrapper around the GW2 API with minimal agent-facing documentation.

• Add 50-150 character descriptions to every tool. Example: 'worlds' → 'Retrieve a list of all Guild Wars 2 game worlds (servers). Optionally filter by comma-separated world IDs. Returns world name, region, language, and population status.'
• Document every parameter with type, format, and constraints. Example: 'ids: Comma-separated world IDs (e.g., "1,2,3"). Omit to retrieve all worlds. Max 200 IDs per request.'
• Add output schema documentation for each tool. Example: 'worlds returns: [{id: number, name: string, region: string, language: string, population: string}]'
• Implement structured error responses with recovery hints. Example: 'HTTP 400: Invalid IDs format. Expected comma-separated numbers (e.g., "1,2,3"). Got: "abc". Try: search_worlds() to discover valid IDs.'
• Add tool annotations to the schema. Mark all tools with readOnlyHint: true since they are safe to cache and retry. Example: inputSchema: {..., readOnlyHint: true}