sparkyfitness-mcp-server
SparkyFitness ships a first-party Model Context Protocol server IN-PROCESS inside the main API server — there is no separate MCP service to run. It is a stateless StreamableHTTP transport mounted at POST /mcp on the operator's own deployment, built on @modelcontextprotocol/sdk, and it exposes the same tool registry the built-in AI assistant uses. Because SparkyFitness is self-hosted, the endpoint host is the operator's, not a vendor's.
One-click install for Cursor, VS Code, Claude, and 20+ other MCP clients, powered by API Commons MCP Install — visit install.apicommons.org for more information.
Documentation
Documentation link · transport streamable-http (stateless; a fresh McpServer + transport per request)
Tools
sparky_execute_read_only_sql— Execute a custom read-only SQL query (SELECT or WITH statements) against Postgres for troubleshooting. Requires admin access and DEV_TOOLS_ENABLED=true.sparky_get_db_stats— Get current database connection pool statistics. Requires admin access and DEV_TOOLS_ENABLED=true.sparky_get_user_info— Get information about the current authenticated user. Requires admin access and DEV_TOOLS_ENABLED=true.sparky_inspect_schema— Inspect the database schema to understand available tables and columns. Requires admin access and DEV_TOOLS_ENABLED=true.sparky_query_table— Execute a read-only query on a database table for troubleshooting. Requires admin access and DEV_TOOLS_ENABLED=true.sparky_get_sleep_science— Read sleep-science analytics: sleep debt (sleep_debt), MCTQ chronotype stats (mctq_stats), daily sleep need for a date (daily_need), circadian energy curve (energy_curve), chronotype summary (chronotype), data sufficiency check (data_sufficiency), and recalculate the baseline sleep need (recalculate_baseline). Mostly read-only; recalculate_baseline persists an updated baseline.sparky_manage_checkin— Health tracking: weight, steps, body measurements, mood, sleep, fasting, custom metrics.sparky_manage_progress_photos— Progress (check-in) photos: list the days a user has progress photos, list the photos for a specific day, and delete a photo.sparky_analyze_trends— Analyze weight trends vs. calorie intake to identify plateaus or progress over a specified number of days.sparky_check_engagement— Scans the user's data for moments that require a proactive nudge (e.g., missed workout, plateau, achievement).sparky_daily_checkin_wizard— A guided, step-by-step interactive assistant for your daily health check-in. Use 'daily_checkin' action and pass the current step.sparky_detect_patterns— Health Detective: Scans historical data for correlations between nutrition, sleep, and mood.sparky_generate_coaching_plan— Auto-Coach: Generates a 7-day macro plan and shopping list based on your goal and weight trends.sparky_get_30_day_trends— Get comprehensive trends for the last 30 days including food, exercise, mood, sleep, and biometrics.sparky_get_contextual_nudge— Generates a context-aware nudge based on recent user activity or inactivity.sparky_get_health_summary— Get a summary of the user's health status (Nutrition, Fitness, Vitals, Hydration) for a specific date range.sparky_get_logging_streak— Retrieves the user's current consecutive logging streak for any health or fitness data.sparky_get_daily_exercise_totals— Returns daily exercise totals for a date or range.sparky_get_exercise_details— Returns full details for one exercise by exercise_id or exercise_name.sparky_get_exercise_diary— Returns entry-level exercise diary data for a specific date or date range.sparky_get_exercise_progress— Returns paginated performance history for an exercise.sparky_get_exercise_stats— Read exercise analytics: aggregated stats over an interval (stats_summary), advanced activity search (query_activities), personal records / best efforts (personal_records), and matched course groupings (matched_courses). Read-only.sparky_get_exercise_usage— Shows where a specific exercise_id was used in the exercise diary.sparky_get_recent_exercise_entries— Returns recent entry-level exercise diary rows for the authenticated user.sparky_list_exercises— Returns a paginated exercise catalog for the authenticated user.sparky_manage_exercise— Fitness tracking: search exercises, log workouts with sets, manage presets.sparky_manage_workout_plans— Workout plan templates: list the user's saved workout plans, inspect one in detail, or delete a plan.sparky_search_exercises— Searches exercises by name and optional filters.sparky_get_barcode— Look up a food product by its barcode (lookup_barcode). Searches your saved foods first, then configured external providers (OpenFoodFacts, FatSecret, etc.), and returns the matched food with per-serving macros. Read-only.sparky_get_food_details— Returns full details for one food by food_id, including available variants.sparky_get_food_diary— Returns entry-level food diary data for a specific date or date range.sparky_get_food_usage— Shows where a specific food_id was used in the diary.sparky_get_nutrition_summary— Returns nutrition summary rows for a specific date or date range.sparky_get_recent_food_entries— Returns recent entry-level food diary rows for the authenticated user.sparky_list_foods— Returns a paginated food catalog for the authenticated user, including variants.sparky_manage_allergens— Allergen preferences: list, add, and remove the allergens a user tracks.sparky_manage_custom_nutrients— Custom nutrients: list, view, create, update, and delete a user's custom nutrient definitions (name + unit + aliases) that layer on top of the built-in nutrient catalog.sparky_manage_favorites— Favorites: list, add, and remove the foods and meals a user has saved as favorites for quick logging.sparky_manage_food—sparky_manage_meal_plans— Meal plan templates: list the user's saved plans, inspect one in detail, duplicate a plan, or delete a plan.sparky_manage_water_containers— Water containers: list, view, create, update, delete, and set the primary reusable water vessel (name + volume + unit) a user taps for quick water logging. Volume is stored in ml; provide the unit (ml, oz, or liter) and it is converted automatically.sparky_search_foods— Searches foods by name for the authenticated user.sparky_get_goal_snapshot— Returns the goals active on a specific date.sparky_manage_goals— Target management: set and view calorie, macro, water, and weight goals.sparky_manage_medications— Medication tracking: manage medications and schedules, log doses, and view history.sparky_ask_user— Asks the user a clarifying question mid-turn before continuing (ASK_USER_TOOL_NAME).sparky_enable_tools— Turns additional tool categories on for the current session (ENABLE_TOOLS_TOOL_NAME).sparky_get_integrations— Read connected external integrations and available integration types. Actions: list_providers (connected external data providers for this user with status, visibility, and sync info — no credentials exposed), list_provider_types (catalog of available integration types). Read-only; does not manage credentials.sparky_get_synced_data— Synced provider data: list the entry data a user has synced from external providers (e.g. garmin, healthkit, health_connect). Read-only.sparky_manage_habits— Habit tracking: list habits, log completions, and view history.sparky_manage_profile— User settings: update display name, timezone, and measurement units.sparky_get_daily_report— Returns daily report data across nutrition, exercise, and water for a specific date or range.sparky_get_dashboard— Read the daily dashboard calorie-balance summary for a date (daily_summary): calories eaten, burned, remaining, goal, net, progress, steps, step calories, and BMR. Defaults to today. Read-only.sparky_get_report— Generates consolidated health and fitness reports.sparky_analyze_food_image— Analyzes an image of food to estimate its nutritional content using advanced vision models.sparky_scan_label— Scans a nutrition label from an image to extract detailed nutritional information using OCR.
About MCP
The Model Context Protocol (MCP) is an open protocol Anthropic introduced for connecting LLM-based agents to external tools and data sources. Providers publish MCP servers that expose their API surface as structured, discoverable tools — an MCP-compatible client (Claude Desktop, Cursor, Cline, Continue, etc.) can connect to the server and call its tools without any per-provider integration code.
Browse every MCP server on the APIs.io network or compare with the broader Agent Skill surfaces of the same providers.
Work with this as data
Every MCP server here is available over the APIs.io API and to AI agents over MCP.