The server defines 3 tools with varying quality. Tool schemas are present and mostly well-formed, but descriptions are minimal (10-60 chars), parameter documentation is sparse, and error handling guidance is absent. The 'test' tool is trivial and does not align with production use. 'analyze_image' and 'analyze_webpage' have actionable schemas but lack crucial details about expected behavior, failure modes, and recovery steps. No output schemas are documented. Per the rubric baseline of 194 chars for tool descriptions, these fall significantly short. Schema coverage is present but parameter descriptions are minimal.

• Expand tool descriptions to 100 - 200 characters. Describe WHAT the tool does, WHEN to use it (e.g., 'Use for image captioning, object detection, or visual question answering'), and WHAT it returns. Example: 'Generate captions, detect objects, or answer questions about images using the Moondream vision model. Input an image path and a prompt (caption, detect: object, or question). Returns analysis text.'
• Replace example-based prompt guidance with formal constraints. Add an enum parameter 'analysis_type' with options ['caption', 'detection', 'vqa'] OR document 'prompt' format as a regex pattern and provide an enum for common prefixes.
• Document output schemas for all tools. Specify: test returns {result: string}; analyze_image returns {analysis: string, type: string, confidence?: number}; analyze_webpage returns {analysis: string, screenshot_path?: string}.
• Add parameter validation rules to descriptions. For 'image_path': 'Absolute or relative filesystem path to a .jpg, .png, .gif, or .webp file (max 100 MB). File must exist and be readable.' For 'url': 'HTTP or HTTPS URL (max 2048 characters). Must be reachable and return HTML.'
• Add error handling guidance to tool descriptions. Example: 'If the file does not exist, returns error 'image_path not found'. If Moondream fails to load, returns 'Model initialization failed, try again later'. If analysis times out after 60 seconds, returns 'Analysis timeout, image may be too large'.'
• Rename tools for clarity: 'analyze_image' → 'generate_image_caption', 'detect_image_objects', 'ask_image_question' (separate tools per intent) OR 'analyze_image' stays but description explicitly lists supported analysis types.