Both tools have minimal descriptions (under 100 chars) and lack comprehensive parameter documentation. gen_text_overlay has 4 parameters with basic type info but no descriptions for blend_mode or out_dir. gen_image has 2 parameters with minimal guidance on valid step ranges. Neither tool documents output schemas, error handling, or recovery paths. No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite both being WRITE operations. Parameter defaults are present but not justified. No evidence of input validation or constraint documentation (e.g., valid blend modes, step bounds).

• Add comprehensive parameter descriptions: for gen_text_overlay, document valid blend_mode values (e.g., 'normal, multiply, screen, overlay') as an enum; clarify out_dir format ('local file path or S3 URI'). For gen_image, specify steps range ('integer, 1 - 50, default 10').
• Document output schemas for both tools. E.g., gen_text_overlay returns {success: bool, output_path: string, error?: string}; gen_image returns {url: string, seed: int, dimensions: {width, height}}.
• Add error handling guidance: 'If Segmind API fails, retry after 5 seconds. If rate-limited, wait 60 seconds. If image generation times out, reduce steps parameter.'
• Declare tool annotations: mark both as destructiveHint=true (they write files/generate resources). Mark gen_image as idempotentHint=false if the same prompt produces different outputs (non-deterministic).
• Expand tool descriptions from ~50 chars to 150 - 200 chars. Include WHEN to use each tool and what it returns. E.g., 'Overlay text on an image using Segmind API. Returns the URL of the processed image. Use this to add captions, watermarks, or labels to existing images.'