Four tools with partial schemas and inconsistent descriptions. extract_edges and generate_from_uin have complete input schemas with proper constraints (min/max, enums, defaults), but descriptions are generic (50-60 chars). analyze_image and validate_uin lack visible schema definitions in the provided code, only inferred from tool names. No output schemas documented. Parameter descriptions are minimal. No error handling guidance. Tool names are action-oriented (extract_, generate_, analyze_, validate_) but lack clarity on when to use each vs. alternatives.

• Add complete input schemas for analyze_image and validate_uin. Specify parameter types, descriptions, constraints (enums, min/max, patterns), and required fields. Currently inferred only from names.
• Document output schemas for all tools. E.g., extract_edges should return {file_path: string, uin_json: object, edge_image_path: string, metadata: object}. This enables LLM planning.
• Expand tool descriptions to 100 - 200 chars. Include: what it does, when to use it (vs. similar tools), prerequisites, and what it returns. E.g., 'Extract Canny edge features from an image and package them as a UIN (Universal Image Notation) JSON file. Use this to prepare images for ComfyUI generation. Returns edge image path and UIN metadata.'
• Add descriptions to all parameters. For analyze_image.detail_level, explain: 'basic: color palette and composition; detailed: adds object detection and style; forensic: includes pixel-level analysis and anomalies.'
• Add error handling guidance to tool descriptions. E.g., 'If the image format is unsupported, try convert_image_format first. If the path is invalid, call list_available_images to discover valid paths.'