github MBombeck/HealthLog v1.1.0
v1.1.0 — AI Insights Overhaul

latest releases: v1.4.49.3, v1.4.49.2, v1.4.49.1...
one month ago

HealthLog v1.1.0 — AI Insights Overhaul

Complete overhaul of the AI-powered health insights system with medical guideline-based analysis and ChatGPT subscription integration.

Highlights

ChatGPT Proxy Integration

  • Insights now run through your ChatGPT Plus/Pro subscription via a local proxy — no separate API billing
  • Admin fallback: global API key (OpenAI/OpenRouter) for users without their own connection
  • Provider abstraction with automatic failover (Codex OAuth → Admin Key → Fallback texts)

Medical Insight Prompts

7 specialized prompts based on current clinical guidelines:

  • Blood Pressure: ESC/ESH 2023 classification, morning stroke risk ladder (J-HOP), pulse pressure, seasonal variation
  • Weight: 5%/10% milestone recognition, plateau detection, body composition divergence, sleep-weight correlation
  • Pulse: Fitness interpretation ladder, 80–100 bpm elevated-risk band, rate-pressure product
  • BMI: Age-adjusted DEGAM classification for 65+, body fat context
  • Medication: Chronotherapy hints, mood-adherence risk prediction, 90-day tracking
  • General Status: Cross-domain synthesis with cardiovascular risk stratification
  • Mood: Bidirectional correlations with vitals and adherence

New Health Metrics

  • Rate-Pressure Product (pulse × systolic BP — myocardial demand)
  • Body composition divergence (weight stable + body fat rising = muscle loss)
  • Mood-adherence risk predictor (mood decline → adherence drop in 2–5 days)
  • Seasonal BP variation (winter vs summer)
  • BP standard deviation + pulse pressure (arterial stiffness)
  • Sleep duration and activity steps (previously ignored)
  • 5 cross-metric Pearson correlations
  • Historical comparison (current 7d vs previous 30d baseline)

New UI Components

  • InsightStatusCard — compact per-metric card with classification indicator and fade-in animation
  • InsightAdvisorCard — premium structured card with findings, correlations, recommendations (ready for integration)
  • Personal advisor tone ("Dein Blutdruck zeigt...") with positive-first pattern
  • Dracula-themed with proper dark/light support

Security

  • Rate limiting on all OAuth and admin endpoints
  • PKCE (S256) + state parameter for OAuth CSRF protection
  • Encrypted token storage at rest (AES-256-GCM)
  • Error message truncation to prevent upstream body leaks
  • 3 security audits applied
  • prefers-reduced-motion support

Breaking Changes

  • openaiKeyEncrypted field removed from User model (replaced by provider abstraction)
  • Settings API now returns codexStatus instead of hasKey
  • Per-metric insight generators return hasProvider instead of hasKey

Don't miss a new HealthLog release

NewReleases is sending notifications on new releases.