fitbit
Provides AI-powered Fitbit insights by fetching user data from the Fitbit Web API. It reads/stores OAuth credentials from /root/clawd/fitbit-config.json and FITBIT_ACCESS_TOKEN, runs python3 scripts/fitbit_api.py/curl commands, and calls https://api.fitbit.com/oauth2/token.
Fitbit Fitness Insights - Clawdbot Skill
🏋️ AI-powered insights from your Fitbit data!
What It Does
Ask your AI assistant questions about your fitness data and get intelligent insights:
- 📊 Activity insights - "How did I do this week?"
- 💓 Heart rate analysis - "What was my heart rate during workouts?"
- 😴 Sleep tracking - "How did I sleep last night?"
- 🏃 Workout analysis - "Did I exercise today?"
- 📈 Trend analysis - "Am I more active on weekdays or weekends?"
- 🎯 Goal tracking - "Did I hit my step goal this week?"
Example Conversations
You: "How did I sleep last night?"
AI: "You slept 7.9 hours with 96% efficiency. Only woke up twice for 2 minutes total. That's excellent sleep quality! 😴"
You: "How active was I this week?"
AI: "You averaged 8,234 steps/day this week (up 12% from last week!). Hit your 10k goal 4 out of 7 days. You logged CrossFit 3 times with 156 very active minutes total. Solid week! 💪"
Features
- Natural language Q&A - Ask questions naturally, get conversational answers
- AI-powered insights - Smart analysis, not just raw numbers
- Trend detection - Spots patterns, week-over-week changes
- No app needed - Works via Fitbit API directly
- Privacy-focused - Read-only access, your data stays private
Installation
-
Install the skill:
clawdbot skills install fitbit.skill -
Set up Fitbit OAuth (takes 5 minutes):
- Register app at https://dev.fitbit.com/apps
- Get OAuth tokens
- Configure
fitbit-config.json
-
Start asking questions!
- "How did I do this week?"
- "Did I exercise today?"
- "What was my sleep like?"
Full setup guide included in the skill package.
What Data You Can Query
- Activity: Steps, distance, calories, active minutes, floors
- Heart Rate: Resting HR, heart rate zones, workout intensity
- Sleep: Duration, efficiency, wake-ups, sleep stages
- Workouts: Logged activities, duration, calories burned
- Trends: Week-over-week comparisons, patterns, achievements
Requirements
- Fitbit account with data syncing
- Fitbit OAuth credentials (free developer account)
- Clawdbot with Telegram or other messaging platform
- Python 3 (included with Clawdbot)
Security & Privacy
- ✅ Read-only access - Never writes to your Fitbit account
- ✅ OAuth 2.0 - Secure token-based authentication
- ✅ Auto-refresh - Tokens refresh automatically
- ✅ No data storage - Queries on-demand, nothing saved
- ✅ Your data - Stays in your Fitbit account, never shared
Setup Time
- 5 minutes for OAuth setup (one-time)
- Auto-refresh - Tokens managed automatically
- Zero maintenance - Just ask questions!
Use Cases
- Track fitness goals and progress
- Analyze workout effectiveness
- Monitor sleep quality
- Compare activity patterns
- Spot correlations (sleep vs performance)
- Get motivated by achievements
Technical Details
- Fitbit Web API integration
- OAuth 2.0 with auto-refresh
- Python 3 (standard library only, no pip packages)
- Config-based token management
- AI-powered insight generation
What's Included
fitbit.skill- Complete skill packageREADME.md- This fileSETUP.md- Step-by-step OAuth setup guide- Inside skill:
- SKILL.md - Full documentation
- fitbit_api.py - API client
- refresh_token.py - Auto-refresh system
- fitbit-oauth-setup.md - OAuth guide
Support
- Issues: https://github.com/clawdbot/clawdbot/issues
- Community: https://discord.com/invite/clawd
- Docs: https://docs.clawd.bot
- Fitbit API: https://dev.fitbit.com/build/reference/web-api/
Ready to get AI-powered fitness insights? 🏋️✨
Download fitbit.skill and start asking questions about your fitness data!
Made with ❤️ for Fitbit users who want smarter fitness tracking