Three tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. All tools have descriptions (10-50 chars baseline), but descriptions are generic and lack actionable context for LLM selection. Input schemas use Zod validation (good), but parameter descriptions are minimal. No output schemas documented. Error handling is generic (validation errors only). Tools follow verb_noun naming but lack depth in composition guidance.

• Document output schemas for all three tools. Example: searchWikipedia returns {title: string, summary: string, url: string, thumbnail?: string}. This enables LLMs to chain results into other tools.
• Expand tool descriptions to 100-200 chars with WHEN/WHY context. E.g., 'Searches Wikipedia for encyclopedic summaries (best for factual topics, not real-time events). Returns article title, 200-word summary, URL, and thumbnail image.'
• Add 50-100 char descriptions to every parameter. E.g., 'Search query (1-300 chars): topic or phrase to look up. Use specific terms (e.g., "photosynthesis") rather than questions (e.g., "how does photosynthesis work").'
• Implement error recovery guidance. On validation failure, return: 'Invalid language code. Supported codes: en, es, fr, de, pl, zh, ja. See https://en.wikipedia.org/wiki/List_of_ISO_639_language_codes'
• Add enum constraints for language parameters or create a list_supported_languages discovery tool. Prevents hallucinated language codes.