This server exposes only 1 tool ('summarize') with a minimal, placeholder implementation. The tool has a basic description (47 chars) and a simple input schema with one string parameter. However, the description is generic and lacks actionable context for LLM selection. The parameter 'text' has a description but no constraints (length, format). The tool's output schema is not documented, callers cannot predict the return structure. The server is clearly a demo/example (note: 'Pretend this calls a model'), not production-grade. No error handling, no pagination, no composition patterns. The fastmcp framework is used but transport is unknown from the provided code snippet.

• Expand tool description to 50 - 200 characters explaining WHAT it does, WHEN to use it, and WHAT it returns. Example: 'Summarize text into a concise overview. Use this to condense long documents, emails, or articles. Returns a single-paragraph summary.'
• Document the output schema explicitly. Specify that the tool returns a string and provide an example: 'Returns: string (e.g., "summary of 1000 characters")'.
• Add constraints to the 'text' parameter: minimum length (e.g., 10 chars), maximum length (e.g., 100,000 chars), and format guidance.
• Implement error handling with actionable messages. If text is empty, return 'Error: text parameter cannot be empty. Provide at least 10 characters.' If text exceeds limits, return 'Error: text exceeds 100,000 character limit.'
• Clarify the tool's actual behavior. If this is a demo, document that it returns a mock summary. If it calls a real model, specify which model and include cost/latency expectations.