Releases: elkimek/get-based
Releases · elkimek/get-based
v1.5.3 — Import & Editing Improvements
- When importing a PDF, you can now exclude individual results and see the lab's reference ranges before confirming
- New import button in the bottom-right corner for quicker access to importing
- Changed a value? A revert button lets you go back to what the lab originally reported
- Use any AI model on OpenRouter by typing its ID directly
- AI discussions now pick up where you left off after refreshing the page
- Automatic backups saved daily and kept for 30 days
- App auto-reloads when a new version is available (no more hard refresh needed)
v1.5.2 — Chat & Discuss Mode Improvements
- Get a second opinion — pick which AI persona joins your chat discussion
- Marker charts and values display correctly in the detail view
- Create custom AI personas with a cleaner editor and auto-generated emoji
- New "Unconventional Views" option when creating AI personas
- Imported vs manually entered values are now labeled in Settings
v1.5.1 — EMF Improvements & Mobile
- EMF: sleeping vs daytime SBM-2015 thresholds — severity adjusts per room type
- EMF: AI interpretation modal — stream analysis of single assessments or before/after comparisons, saved with assessment
- EMF: meter presets with autocomplete, room photos (up to 6 per room), before/after comparison view
- Mobile: hamburger menu with slide-out sidebar replaces stacked category pills
- Mobile: cleaner header — hides dates, range toggle, feedback, and donate on small screens
- Desktop: header groups with subtle dividers between profile, data, and actions
v1.5.0 — EMF Assessment
- Environment card: Baubiologie EMF assessment sub-module — track electromagnetic field measurements room by room
- SBM-2015 severity grading (No Concern → Extreme Concern) for 5 measurement types: AC electric, AC magnetic, RF/microwave, dirty electricity, DC magnetic
- Import EMF consultant reports via PDF — AI extracts rooms, measurements, sources, and mitigations
- Printable consultant template for on-site assessments (download from EMF editor)
- EMF data included in AI chat context, JSON export/import, and environment summary
v1.4.3 — Improved Fatty Acids Support
- Improved support for fatty acid panels — each lab appears as its own subcategory under a "Fatty Acids" sidebar group
- Auto-detects fatty acid lab from PDF content
- Re-importing a PDF now updates category labels instead of keeping stale ones from previous imports
v1.4.2 — Bugfixes & Improvements
- Edit any value — click a value in the detail modal to change it inline, with "edited" badge
- Fix: manually added values now store correctly in US unit mode
- Compact drop zone on dashboard after first import
- Auto-backup cooldown increased to 5 minutes to avoid snapshot churn
- PII review: retry button stays visible after successful obfuscation
v1.4.1 — Bugfixes & Improvements
- Fix: lab reference ranges from US-unit PDFs now import correctly
- Fix: PII review diff highlighting works properly with thinking models
- Improved PII review editing — cursor lands where you click, highlights persist after edits
- Better matching for US lab markers like BUN
v1.4.0 — Image Attachments
- Chat: attach images via paperclip button, paste, or drag-and-drop — AI can see photos of lab reports, supplement labels, food logs, and more
- Chat: up to 5 images per message, resized for optimal quality vs token cost
- PDF import: scanned/image-heavy PDFs now detected automatically with image mode fallback — renders pages as screenshots for AI analysis
- PDF import: "Force image mode" link in drop zone for known scans
- Vision support detection: attach button auto-hidden for non-vision models
- HD mode toggle: switch between standard (1024px) and high-resolution (2048px) image quality
- Image quality warnings: detects blurry, dark, or overexposed photos before sending
- All image metadata (EXIF, GPS, camera info) automatically stripped before sending
- Click-to-expand lightbox for chat image thumbnails
- Chat: smooth typewriter streaming replaces chunky text updates
- Chat: conversation window expanded from 10 to 30 messages for better context
v1.3.7 — Reference Ranges & PII Improvements
- Editable reference ranges — click any range in the detail modal to customize, with revert badge
- Import-time range adoption — toggle to use your lab's reference ranges from the PDF
- BUN/Creatinine ratio added as a calculated marker in the Kidney category
- PII review: green word-level diff highlighting on the right panel, click-to-edit
- PII review: model unloads from VRAM on Stop, Cancel, or Use Regex
- US lab PII patterns: Specimen ID, Accession No, Account No, MRN, phone, Member ID
- IU/L enzyme unit normalization (ALT, AST, ALP) — no more double-conversion on import
- Empty marker charts hidden in category views
- Background scroll locked on PII review modals
v1.3.6 — Settings Cleanup
- Profile tab removed from Settings — sex, date of birth, and location are now in the Client List
- Location field in Client List now shows live latitude with AI refinement (same as the old Settings)