AI Features
The project combines deterministic safety rules, provider-backed generation, and MCP tools so AI output stays explainable and resilient.
Key Points
Rule-based fallback protects the product when AI providers fail.
Groq and Gemini can generate richer explanations when keys are configured.
MCP tools expose reusable health, nutrition, food, clinic, and exercise workflows.
Risk Analysis
Risk scoring checks health profile and vital signals such as blood pressure, fever, blood sugar, BMI, symptoms, and existing conditions.
Returns low, medium, or high risk levels.
Shows reasons instead of opaque scores.
Escalates dangerous symptoms toward medical care.
Diet And Exercise Planning
The app builds budget-aware Bangladeshi diet plans and weekly exercise plans. It can fall back to local rules if provider output is unavailable or invalid.
Diet planning uses budget, preferred foods, avoided foods, and health profile.
Exercise planning considers risk level and safety warnings.
Responses are shape-validated before being shown to users.
MCP Tooling
The MCP server gives the AI structured tools such as food nutrition search, nearby hospital search, BMI calculation, calorie estimation, and medical-report assistance.