macrofactor-mcp
MacroFactor MCP is a local Fueling & Recovery Decision Engine. Give your MCP client a MacroFactor export, optionally add Garmin, and ask for a decision instead of manually comparing nutrition, sleep, body, and training screens.
README
Am I recovered? Did my fueling work? What should change next week? The only MCP combining MacroFactor nutrition with Garmin training data.
MacroFactor MCP is a local Fueling & Recovery Decision Engine. Give your MCP client a MacroFactor export, optionally add Garmin, and ask for a decision instead of manually comparing nutrition, sleep, body, and training screens.
Quickstart
Install uv, then confirm the packaged server starts:
uvx macrofactor-mcp
Export your data from MacroFactor, put the .xlsx file in one local directory,
and add this to your MCP client configuration. Use absolute paths.
{
"mcpServers": {
"macrofactor": {
"command": "uvx",
"args": ["macrofactor-mcp"],
"env": {
"MACROFACTOR_DATA_DIR": "/absolute/path/to/macrofactor-exports",
"MACROFACTOR_DB_PATH": "/absolute/path/to/macrofactor.duckdb"
}
}
}
}
Restart the MCP client. The server imports new workbooks on startup. Run
setup_check, then ask one of the questions below. If a desktop client cannot
find uvx, replace "uvx" with the absolute path returned by which uvx.
Four flagship workflows
All examples below use synthetic data.
recovery_check — Am I recovered?
Combines last night's sleep, HRV, body battery, training load, and yesterday's fueling into a readiness view.
Recovery Check: 2026-07-08
Sleep: 7h 38m, score 82; HRV +4 vs 7-day average
Training load: ACWR 1.08 (optimal)
Yesterday: 96% calorie target; 101% protein target
Overall: All systems go — 4 positive signals.
daily_briefing — What do today's signals say together?
Shows the day as one picture instead of separate nutrition and training logs.
Daily Briefing: 2026-07-08
Intake: 2,340 kcal (P 171g / F 72g / C 248g)
Adherence: 98% calories | 101% protein
Body Battery: 28 → 83 | Sleep: 7h 38m, score 82
Training: Productive | Load 612/568 (ACWR 1.08)
weekly_report — What should change next week?
Reviews seven days of nutrition, weight, workouts, sleep, and training load.
Weekly Report: 2026-07-02 to 2026-07-08
Average: 2,315 kcal | 168g protein
On target: calories 6/7 days | protein 7/7 days
Training: 4 sessions | Sleep: 7h 26m average
Next-week signal: fueling was consistent; protect sleep before adding load.
nutrition_performance_correlation — Did my fueling work?
Compares nutrition with next-day Garmin recovery signals. It reports sample sizes so small or missing datasets stay visible.
Nutrition → Performance Correlation: 28 paired days
Protein >=90% target: next-day HRV 51 vs 45 (+6)
Protein >=90% target: sleep score 81 vs 76 (+5)
Energy deficit days: body battery 68 vs 75 (-7)
Observation: higher-protein days tended to precede better recovery.
MacroFactor-only or MacroFactor + Garmin
| Mode | Decisions available |
|---|---|
| MacroFactor only | Nutrition, weight, workouts, daily_briefing, and weekly_report. Recovery output uses the nutrition signals that exist. |
| MacroFactor + Garmin | All four flagship workflows, including sleep, HRV, body battery, activities, and training load. |
Garmin is optional. To enable it, point GARMINTOKENS at an existing local
Garth/Garmin token directory, then run sync_garmin.
Platform support
- macOS and Linux
- Windows is untested
- Python 3.10 or newer
Limitations
- v0.1 requires a manual MacroFactor export; it does not connect to a MacroFactor API.
- Garmin correlation workflows need overlapping MacroFactor and Garmin dates.
- Correlations are observational, not proof that one behavior caused another outcome.
- Import support follows the current MacroFactor Quick Export and All-Time Data workbook layouts.
Privacy
Raw nutrition, body, workout, and Garmin data stays in the local DuckDB database. The server does not upload that database. However, MCP tool results are sent to whatever model or client the server is connected to. Review that client's data policy and avoid requesting more detail than you want to share.
Disclaimer
This is an unofficial project. It is not affiliated with, endorsed by, or supported by MacroFactor or Stronger By Science. It is not medical advice and must not replace care from a qualified health professional.
Released under the MIT License.
mcp-name: io.github.NasserAlbusaidi/macrofactor-mcp
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.