Using a local iPhone MCP server to plan Apple Watch workouts with Codex
The article describes how to set up a local MCP (Model Context Protocol) server on an iPhone to connect Apple Health and Workout data with Anthropic's Codex assistant, enabling automated workout planning and analysis based on personal fitness history.
Background
- MCP (Model Context Protocol) is an open standard by Anthropic that lets AI assistants interact with local tools and data sources. This post uses a local MCP server on an iPhone to give Claude access to Apple Health and Workout data.
- Codex is a third-party iOS app that acts as an MCP server host on iPhone, bridging Apple's HealthKit API with AI models like Claude.
- The author sets up a system where Claude reads their past workout data (via HealthKit through Codex) and generates personalized, context-aware training plans — all processed on-device, no cloud upload.
- This matters because it's a concrete example of "AI agents" operating on personal, private data directly on a phone, without sending sensitive health information to external servers.