Coaching and Trends Tools

The coaching tools are designed to provide high-level insights into a user's health status and long-term trends. It aggregates data from various sources (food, exercise, check-ins) to offer summaries and identify patterns relevant for health coaching.

Tool Names: sparky_get_health_summary, sparky_analyze_trends, sparky_get_30_day_trends, sparky_detect_patterns, sparky_generate_coaching_plan, sparky_review_calorie_target (a group of separate tools, not a single one)

Tools within Coach

sparky_get_health_summary

• Description: Get a summary of the user's health status (Nutrition, Fitness, Vitals) for a specific date range.
• Parameters:
  • start_date (string, YYYY-MM-DD): Start date for the summary.
  • end_date (string, YYYY-MM-DD, optional): End date for the summary. Defaults to start_date if not provided.

sparky_analyze_trends

• Description: Analyze weight trends vs. calorie intake to identify plateaus or progress. (Feature coming in full Phase 3 implementation!)
• Parameters:
  • days (number, optional): Number of days to analyze. Defaults to 7.

sparky_review_calorie_target

• Description: Read-only weekly check of the saved calorie target. Uses the adaptive TDEE (services/AdaptiveTdeeService.ts) as maintenance, the rate the target implies against the rate the weight trend shows, average intake against the target, and the calories for common weekly rates or a chosen one. Judges nothing while the estimate is still learning or fewer than 70% of the days were logged, flags targets under the clinical minimum, and tells the model not to lower a target the user did not follow. The arithmetic is reviewCalorieTarget in shared/src/utils/calorieTargetReview.ts. Changing the goal is sparky_manage_goals set_goals.
• Parameters:
  • desired_kg_per_week (number, -1 to 1, optional): Weekly change the user wants; negative to lose.

sparky_get_30_day_trends

• Description: Get comprehensive trends for the last 30 days including food, exercise, mood, sleep, and biometrics. This tool gathers aggregated data across these domains to provide a holistic view of the user's progress and patterns.
• Parameters:
  • end_date (string, YYYY-MM-DD, optional): End date for the 30-day period. Defaults to today.