This server has 3 tools with schemas present but significant quality gaps. Tool descriptions are present and reasonably detailed (60-130 chars each), but parameter descriptions are minimal or missing. fetch_url and fetch_image have single required params with adequate descriptions; fetch_pdf has 3 params with mixed quality. No output schemas are documented. Parameter type definitions are present (string, enum, array) but descriptions are sparse. The server lacks error handling guidance, recovery suggestions, and output documentation. Naming is reasonable (verb_noun pattern) but tool composition suggests potential overlap, fetch_url, fetch_image, and fetch_pdf appear to be related extraction operations that might benefit from a unified interface or clearer distinction. The schemas are valid JSON Schema but lack the depth needed for production LLM integration.

• Document output schemas for all three tools. Example for fetch_url: '{"type": "object", "properties": {"title": {"type": "string"}, "content": {"type": "string"}, "error": {"type": "string"}}, "required": ["content"]}', helps LLMs plan downstream operations.
• Expand fetch_pdf parameter 'pages' description to: 'Optional array of page numbers or ranges (e.g., [1, "3-5", 7]). Strings denote ranges; numbers are individual pages. If omitted, all pages extracted up to server render limit (typically 100). Ranges are inclusive.'
• Add error handling guidance to each tool description. Example for fetch_url: 'Returns error field if URL is invalid, unreachable, or page fails to render. If error occurs, try alternative URL or check URL syntax. Timeouts after 60s, consider splitting large pages.'
• Clarify tool composition: rename fetch_url → extract_page_content (returns markdown/text); rename fetch_image → fetch_image_binary (returns image bytes, not markdown); rename fetch_pdf → extract_pdf_content (returns text or image array). Or unify under a single extract_url tool with a content_type parameter.
• Shorten fetch_image description from 130 chars to ~80 chars core action; move URL resolution logic to parameter description: 'URL of image to fetch. For site-relative links (e.g., /images/foo.png), construct full URL first: https://example.com/images/foo.png.'
• Add a 'mode' or 'return_format' parameter to fetch_url to control output (markdown, HTML, plaintext), gives LLMs explicit control and reduces ambiguity.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry