Two tools with minimal documentation. Both have basic descriptions (50-60 chars, below the 194-char baseline) and input schemas present but sparse. search_engine lacks enum constraint for 'engine' parameter despite listing valid values in description. scrape_as_markdown has only one required parameter with no output schema documented. No error handling guidance, no parameter validation rules, no recovery hints. Tool names are acceptable (verb_noun pattern) but descriptions are too terse to guide LLM selection effectively. No tool annotations (readOnlyHint, etc.) despite both being read-only. Schemas lack depth, no min/max bounds, no format specifications, no chaining IDs in responses.

• Expand search_engine description to 150-200 chars: 'Search Google, Bing, or Yandex for current information. Returns organic results with URLs, titles, and snippets. Use when you need real-time web data or recent news. Limit results to 10 per query for efficiency.'
• Expand scrape_as_markdown description: 'Fetch and convert a webpage to markdown format. Returns page content, title, and metadata. Use after search_engine to extract detailed information from selected URLs. May fail if page requires authentication or blocks bots.'
• Convert search_engine 'engine' parameter to enum: {"type": "string", "enum": ["google", "bing", "yandex"], "default": "google", "description": "Search engine to use. Defaults to Google."}
• Document output schemas: search_engine returns {"organic": [{"link": string, "title": string, "snippet": string}], "total_results": number}. scrape_as_markdown returns {"content": string, "title": string, "url": string}.
• Add parameter constraints: search_engine 'query' should have minLength: 1, maxLength: 2048. Add error guidance: 'If no results, try simpler keywords or different engine.'