Multi-device athletes
Ask questions across training, sleep, HRV, recovery, and workout history from more than one platform.
Independent tool overview
freddy at a glance
freddy is a personal health-data hub and MCP server that lets compatible AI assistants query the fitness, sleep, recovery, glucose, and environmental metrics a user chooses to connect.
Visit the official freddy site ↗
- Best for
- Cross-source fitness, sleep, recovery, glucose, and health-data analysis
- Free plan
- One connected source and seven days of history
- Pro price
- $49 per year
- Interfaces
- MCP, ChatGPT app, iOS, Android, dashboard, CLI, and Personal API
- Age requirement
- 18 or older
- Last reviewed
- August 29, 2026
Overview
What freddy is
freddy connects wearable, fitness, glucose, strength-training, and environmental data to ChatGPT, Claude, Perplexity, coding agents, and other MCP-compatible clients. Instead of providing its own chatbot, it normalizes and stores data from connected sources, then serves the relevant metrics to the AI assistant the user chooses.
The main advantage is cross-source analysis. An athlete can compare training load with sleep and HRV, while a biohacker can examine glucose alongside sleep or air-quality readings. Apple Health and Health Connect also provide a route for many apps that do not have a direct freddy connector.
This is sensitive-data infrastructure, not medical care. freddy stores a copy of connected health data in the United States and passes requested metrics to the selected AI client. Its terms state that the service is not a medical device and should not be used to make medical decisions.
Use cases
Who freddy is best for
The strongest fit depends on the job you need the product to complete, not the size of its feature list.
Self-tracking experiments
Compare signals such as glucose, sleep, nutrition, training load, and home air quality over time.
Long-term trend review
Use Pro's full-history access to examine changes in resting heart rate, weight, sleep, or performance across longer periods.
AI-agent builders
Expose personal health metrics to MCP clients or use the read-only Personal API in private dashboards, scripts, and automations.
ChatGPT and Claude users
Bring current device data into an existing AI workflow instead of adopting another closed coaching chatbot.
Capabilities
Core freddy features
Personal MCP endpoint
One private MCP connection can serve normalized health metrics to compatible assistants and agents.
Broad source support
Live connectors include Apple Health, Health Connect, Garmin, Oura, Polar, Suunto, Ultrahuman, Wahoo, WHOOP, Withings, Airthings, Concept2, Hevy, and Intervals.icu, with additional beta and planned sources.
Mobile health-platform sync
The iPhone app reads authorized Apple Health data, while the Android app reads authorized Health Connect data and can sync in the background.
Cross-source questions
Ask an AI to compare metrics across providers, such as recent training load against sleep duration and HRV.
ChatGPT app
freddy is listed as a ChatGPT app and can be invoked from a conversation after the account and sources are connected.
Audit and revocation controls
The dashboard shows a per-request audit log, and users can revoke an AI client, disconnect a source, rotate access, or delete the account.
CSV export
Users can download their stored history for use outside freddy.
Personal API
Pro and Believer accounts can create read-only API keys for metrics, catalogs, source lists, and profile data. The API is currently beta.
Agent and CLI support
A command-line workflow supports headless agents, while MCP account tools can manage connection, disconnection, and sync actions.
Process
How the freddy workflow works
Step 1
Create and secure the account
Sign in with the emailed code, review the privacy terms, and treat the personal MCP URL like a password because anyone holding it can query the account.
Step 2
Connect only needed sources
Authorize selected wearables or apps with read access, or use Apple Health and Health Connect as an aggregation layer.
Step 3
Connect a trusted AI client
Add freddy to ChatGPT, Claude, Perplexity, or another MCP client after reviewing that client's own data-use and retention settings.
Step 4
Ask bounded questions
Specify the metrics, source, and date range, and ask the assistant to separate raw observations from interpretation.
Step 5
Verify the underlying data
Compare important conclusions with the source app, account for device differences and missing sync periods, and do not treat AI interpretation as medical advice.
Step 6
Audit and disconnect
Review the request log, revoke unused clients, rotate exposed tokens, and disconnect sources or delete the account when access is no longer needed.
Cost
freddy pricing and free plan
freddy has a usable free tier, a $49 annual Pro subscription, and a $249 one-time early-supporter option. The AI assistant connected to freddy may require its own separate plan.
Free
$0
A limited personal connection for trying the MCP workflow.
- One connected source
- Seven days of history
- Works with compatible MCP chats and agents
- No payment card required
freddy Pro
$49 per year
Full-history and multi-source access.
- Unlimited connected sources
- Full history for every available metric
- Cross-source analysis
- Early access to new connectors
- Personal API beta access
Believer lifetime
$249 one time
A non-recurring early-supporter purchase with ongoing Pro features.
- Lifetime Pro access
- Includes future connectors presented as part of Pro
- Sold on the freddy website
- Offer availability should be confirmed before purchase
Pricing checked . Check current pricing at the source ↗
Assessment
freddy strengths and limitations
Where it stands out
- Combines data from multiple health and fitness ecosystems behind one interface
- Works with the AI client and model the user already prefers
- Provides a meaningful free plan and unusually simple $49 annual Pro price
- Uses read-only provider permissions and does not write back to connected wearable accounts
- Supports Apple Health and Health Connect, expanding coverage beyond direct integrations
- Provides account deletion, per-source disconnection, CSV export, token revocation, and an audit log
- States that freddy does not sell health data or use stored health metrics to train models
- Offers MCP, a listed ChatGPT app, mobile apps, CLI access, and a read-only Personal API
What to consider
- freddy stores a copy of highly sensitive health data, potentially including precise activity routes, medication, reproductive, mental-wellbeing, and glucose information
- Stored data is handed to the AI client selected by the user, and that client's privacy, retention, and training rules apply separately
- Anyone with the personal MCP URL can query the account, so accidental exposure requires immediate revocation or rotation
- The service is not a medical device and does not provide medical advice; wearable data and AI interpretation should not drive medical decisions
- Provider measurements can disagree because devices use different sensors, sampling methods, definitions, and proprietary scores
- AI answers can overstate correlations, miss missing-data periods, or confuse provider-derived scores with raw measurements
- Data is stored on infrastructure in the United States, while edge processing may occur in other locations; UK and EEA users should review the explicit transfer disclosures
- The free plan's one-source and seven-day limits prevent meaningful cross-source or long-term analysis
- Connector availability and completeness depend on third-party APIs, mobile background sync, and each provider's history window
- Some sources are beta, alpha, indirect through Apple Health or Health Connect, or only planned
- The Personal API is beta and is limited to the account's plan window, 120 requests per minute per key, and up to 10 active keys
- freddy is intended only for adults age 18 and older
Compare
freddy alternatives
The right alternative depends on the specific output, workflow, controls and budget your project requires.
Healthcare
Health in ChatGPT
Consider ChatGPT Health when you want a dedicated ChatGPT health experience with supported health-record and wellness connections rather than a portable MCP layer.
Explore Health in ChatGPT →Healthcare
Docus
Consider Docus when the primary need is an AI health assistant and clinician-oriented second-opinion workflow rather than wearable-data aggregation.
Explore Docus →Healthcare
Empirical Health
Consider Empirical Health when you want clinician-backed preventive care rather than self-directed analysis of wearable metrics.
Explore Empirical Health →Questions
freddy FAQs
What is freddy?
freddy is a personal MCP server that syncs authorized health, fitness, and environmental data and makes it queryable by a compatible AI assistant.
Which devices and apps does freddy support?
Current live sources include Apple Health, Health Connect, Garmin, Oura, Polar, Suunto, Ultrahuman, Wahoo, WHOOP, Withings, Airthings, Concept2, Hevy, and Intervals.icu. Some other sources are beta, alpha, indirect, or planned.
How much does freddy cost?
Free includes one source and seven days of history. Pro costs $49 per year for unlimited sources and full history. A $249 one-time Believer lifetime offer was also listed when reviewed.
Does freddy include an AI assistant?
No. freddy stores and serves the data; ChatGPT, Claude, Perplexity, or another compatible client performs the AI processing and may have separate pricing.
Does freddy train on health data?
freddy states that it does not perform AI processing on stored health metrics, train models on them, sell them, or use them for advertising. The connected AI client has its own data policy.
Is freddy read-only?
Its provider credentials and Personal API are read-only, so it does not write health or workout data back to connected services. Its MCP surface can still perform account actions such as connecting, disconnecting, or syncing a provider when the user requests them.
Can freddy provide medical advice?
No. Its terms say it is not a medical device and does not provide medical advice. Consult a qualified healthcare professional for medical decisions.
What happens when I disconnect a source?
freddy states that disconnecting a provider permanently deletes the data synced from that provider. Deleting the account removes the account, connectors, stored health metrics, and access token.
Where is freddy data stored?
freddy's privacy policy says account data is stored on United States infrastructure, with database, transport, and additional application-layer encryption described in the policy.
Bottom line
Our freddy verdict
freddy is a compelling option for technically comfortable athletes and self-trackers who want one conversational layer across several data sources. Its $49 annual Pro tier is straightforward, and the portable MCP model avoids locking analysis to one chatbot. The tradeoff is significant: users are centralizing sensitive health data and exposing it to a chosen AI client. Protecting the MCP credential, reviewing both companies' privacy settings, verifying source measurements, and keeping medical decisions with qualified professionals are essential.
Visit freddy website ↗Similar Consumer Tools
View all Consumer AI tools →
Claude Fable
Claude Fable - Anthropic's new frontier Mythos-class model with top performance across benchmarks
Fusion
Fusion - OpenRouter's new tool that fuses several models into panels to achieve frontier performance
NVIDIA Nemotron 3 Ultra
Nemotron 3 Ultra - Nvidia’s open 550B reasoning model for agents
Framer 3.0
Framer 3.0 - Canvas-native agents that build and run your whole site
AI training for the future of work.
Get access to all our AI courses, hundreds of real-world AI use cases, live expert-led workshops, an exclusive network of AI early adopters, and more.
AI Courses
Get unlimited access to all of our current & upcoming industry-specific AI courses for the duration of your subscription.
Daily Guides
To keep up with the rapid pace of AI, our team publishes AI implementation guides daily. Our library contains 300+ practical use cases to automate real-world work.
Workshops
Join weekly, live, interactive sessions with industry leaders who are at the forefront of AI for hands-on implementation guidance and exclusive insights.
Community
Network with an exclusive community of AI-first professionals who are working smarter with AI. Learn how early adopters are using AI in their work and businesses.