Single tool 'calculate_bmi' has basic definition structure but significant gaps in LLM-readiness. The tool name lacks a leading action verb (should be 'get_bmi' or 'calculate_bmi' is marginally acceptable). Description is adequate at 57 characters but lacks guidance on when/why to use it or what the returned BMI value means. Input schema is properly typed (number) with parameter descriptions, but output schema is completely undocumented, LLM has no guidance on interpreting the returned text. No error handling, validation ranges, or recovery guidance. Tool sits at the 52nd percentile of community servers: definitions exist but lack production rigor.

• Add output schema documentation: explicitly define return structure as {bmi: number, category: 'underweight'|'normal'|'overweight'|'obese', description: string}. LLMs need to know what fields to expect to plan downstream actions.
• Add input parameter constraints in descriptions: 'height (integer, 50-250 cm)' and 'weight (number, 1-500 kg)'. Replace free-form numbers with bounded, validated inputs.
• Expand tool description to 100-150 characters with context: 'Calculate Body Mass Index (weight in kg / height in meters squared) to assess if weight is in a healthy range for height. Returns BMI value and health category (underweight/normal/overweight/obese).'
• Implement input validation: reject height < 50 or > 250, weight < 1 or > 500. Return structured error: {error: 'Invalid input', details: 'height must be 50-250 cm', guidance: 'Please provide height in centimeters.'}
• Add BMI interpretation to output: instead of raw number, return {bmi: 23.5, category: 'normal', message: 'Your BMI is in the normal range (18.5-24.9).'} so LLM understands the meaning without post-processing.