B — Some risk signals Graded October 16, 2026Minor concerns worth a quick check.Based on public adoption and transparency signals as of October 16, 2026. Our opinion, not a code audit or security guarantee.
App connector · Data & databases, Cloud & infrastructure
apple-health-mcp
Local-first Apple Health MCP server. Lets AI assistants answer questions about your sleep, workouts, activity and heart data from a local export.
Bottom line: Signal-based grade (4/7 points): well starred on GitHub, listed in several sources, public source code. This reflects adoption and transparency, not a code audit.
- Popularity score
- 48/100
- Rank
- Not ranked: outside the 500 most popular tools
- GitHub stars
- 574
What it does
apple-health-mcp is a local-first MCP server that connects AI assistants to Apple Health data exported from your device, enabling natural-language queries about sleep, workouts, activity, and heart metrics. All data processing happens locally on your machine without sending health information to external servers.
Who it’s for
Health-conscious professionals and builders who want to ask AI questions about their fitness and wellness trends while keeping sensitive biometric data private. Best suited for users comfortable with local data exports and technical setup.
Copy-paste setup
Check these steps against the project’s own instructions before running anything.
Verify against the repo README at https://github.com/neiltron/apple-health-mcp (opens external site) for the latest instructions. Typical setup involves exporting your Apple Health data locally and configuring the MCP server in your AI client (e.g., Claude Desktop) with a JSON entry pointing to the server executable and data path.
Risks
- Local data exposure: Exported Apple Health files contain sensitive biometric data; if your machine is compromised, attackers gain access to complete health history.
- Prompt injection: Malicious prompts could potentially manipulate AI interpretation of health data or trigger unintended data exports.
- Data retention: Unclear how long exported data persists and whether intermediate caches are securely wiped.
- Limited audit trail: Local-first approach may lack logging of which queries accessed which data points.
- Supply chain trust: Single-developer project with moderate GitHub engagement; future maintenance or security updates not guaranteed.
How to stay secure
Least-privilege permissions
- Run the MCP connector in a dedicated, isolated user account with minimal OS permissions; do not use your primary user account
- Restrict file system access to only the Apple Health export directory; use OS-level ACLs to prevent the connector from reading other sensitive files (credentials, SSH keys, browser data)
- Disable network access at the OS level unless the connector genuinely requires external connectivity; use firewall rules to block outbound traffic by default
Network isolation
- Do not expose the MCP connector to the network; only run it locally via localhost/127.0.0.1
- If integration with a remote AI assistant is required, route traffic through a trusted VPN or air-gapped bridge with explicit request/response validation
- Never connect to untrusted AI systems or cloud services; keep health data processing within your control
Monitoring
- Enable application-level logging for all data access queries and export operations; log timestamps, query content, and data fields returned
- Regularly audit exported Apple Health files for unexpected modifications or access patterns
- Monitor process activity and file handle usage to detect unauthorized reads of the health export directory
Red flags to watch for
- Unexpected network connections or DNS lookups initiated by the connector process
- Queries containing suspicious payloads, unusual syntax, or requests to export full datasets without clear health-related purpose
- Exported files growing in size or modification timestamps not matching your explicit exports; indicates possible caching or logging
- Lack of updates or security responses from the maintainer; consider freezing the version and conducting periodic source code review
Alternatives
- Apple HealthKit REST APIs with your own backend (more control, higher setup burden)
- Health data aggregators like Apple Health's built-in Siri integration (less flexible, Apple-managed)
- Manual CSV export + local LLM analysis (maximum privacy, manual workflow)
- Third-party health dashboards with MCP support (easier setup, less privacy-focused)