Server has 9 tools with basic schemas and descriptions in Chinese. All tools have input schemas with type definitions and parameter descriptions. However, descriptions are minimal (10-50 chars), lack actionable context for LLM selection, and omit critical details like when to use each tool vs. alternatives. Output schemas are not documented. Error handling returns generic status/error fields without recovery guidance. Tool naming follows verb_noun pattern (translate_*, get_*, list_*, switch_*, release_*) but lacks clarity on distinctions (translate_text vs. translate_with_terms vs. translate_batch). No tool annotations (readOnlyHint/destructiveHint). Parameters like 'temperature', 'top_p', 'top_k' lack range constraints in descriptions despite numeric types. The 'terms' parameter in translate_with_terms accepts a dict but provides no schema for its structure.

• Expand tool descriptions to 100-200 chars. Example: 'translate_text: Translate text to a target language. Use this for single texts; use translate_batch for multiple texts. Returns translated text, elapsed time, and chunk info if auto_split=true.'
• Document output schemas for all tools. Example: 'Returns {"status": "success"|"error", "result": string, "elapsed_ms": int, "chunks": [string]}'
• Add JSON Schema constraints: temperature {"type": "number", "minimum": 0, "maximum": 2}, top_p {"minimum": 0, "maximum": 1}, top_k {"type": "integer", "minimum": 1, "maximum": 100}
• Define 'terms' parameter schema as {"type": "object", "additionalProperties": {"type": "string"}, "description": "Terminology mapping: source term → target term. E.g. {\"API\": \"接口\"}"}, avoid example values in description text.
• Enhance error responses with recovery guidance: 'Invalid target_lang: got "xyz". Supported: zh, en, ja, ko, fr, de, es, ru, ar. Call list_languages() to see all 38 options.'