Lalabase
Tickets, Doku, Zeiterfassung und Kundengespräche an einem Ort, durchsuchbar per KI. Für Softwarefirmen mit Kundenprojekten. Open Source: auf deinem eigenen Server oder in unserer Cloud in Deutschland.
Translation pending
Dispatch missions, verify mileage with odometer photos, and keep drivers, vehicles and renewals in one place.
BusinessJavaScript · KotlinGitHub @rjsvillahermosa-byteOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
No updates detected yet. GitHub releases will appear here.
Owners of public repositories under a personal GitHub account can sign in to verify ownership and manage the introduction themselves. No administrator approval is needed.
Sign in, generate a verification file, and commit it to this repository’s default branch within 30 minutes. Files on other branches or forks do not count. Then check the file below to manage the project without administrator approval. Keep the file to retain your access.
If needed, you will go to sign-in. After signing in, return here and press your chosen verification button again.
Run /nomorevibe verify in your project folder to verify ownership and manage the introduction yourself.
Verify in one minuteAsk about other verification options
Hotel transport management: missions, AI ODO capture, fuel audit, PMS renewals, voice assistant, custom reports
Translation pending
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
A local, dependency-free-backend rebuild of https://fleet-madison.base44.app
("FleetFlow" / branded TransportLog) — an intelligent hotel transport management
system that automates mileage tracking, driver mission dispatch, and ODO verification
via AI-powered photo extraction.
Built as a faithful duplicate of the live app's UI, data model, and flows — with the
Base44 cloud replaced by a localStorage mock backend so it runs 100% offline.
Open the printed URL (default http://localhost:5173).Read the full README ↗A local, dependency-free-backend rebuild of https://fleet-madison.base44.app ("FleetFlow" / branded TransportLog) — an intelligent hotel transport management system that automates mileage tracking, driver mission dispatch, and ODO verification via AI-powered photo extraction.
Built as a faithful duplicate of the live app's UI, data model, and flows — with the Base44 cloud replaced by a localStorage mock backend so it runs 100% offline.
npm install
npm run dev
Open the printed URL (default http://localhost:5173).
The repo is deploy-ready: npm run build → dist/, served by the zero-dependency
server.js (npm start, SPA fallback, respects $PORT).
Spaceship / Shipmate (spaceship.run → shipmate.run) — git-connected continuous delivery; it builds the container for you (no Dockerfile needed):
git remote add origin <your-github-repo-url>
git push -u origin main
Then connect the repo in the Shipmate dashboard — every push deploys automatically.
Anywhere else (one command each):
npx vercel deploy --prodnpx netlify deploy --prod --dir=dist (or drag ../fleetflow-dist.zip into app.netlify.com/drop)npm start (they run npm install && npm run build first)dist/ (zip at ../fleetflow-dist.zip)Mobile: responsive at 390px+ (audited page-by-page), installable as a PWA
(manifest.webmanifest + logo.svg, standalone display, brand-aware status bar via
theme-color), camera capture inputs (capture="environment") for ODO/receipt/license photos.
The admin can match the hotel/company palette at runtime: 8 presets (Deep Teal, Hotel
Blue, Ocean, Burgundy, Royal Purple, Forest, Sunset, Charcoal) or custom primary +
accent color pickers with live preview. Colors flow through CSS variables
(src/lib/theme.js), so buttons, nav, chips, status colors, the route map, charts and
the phone status bar all follow — persisted per device in localStorage.
Demo login: [email protected] / demo1234 — or click Continue with Google
(simulated). All data is seeded on first boot and lives in your browser only.
Skinned in the warm car-sharing design language (owner's reference board): cream #FAF3E7
canvas with organic mint #CFE3D6 blobs, deep teal #1E7A5A primary (uppercase Poppins
buttons, active nav, brand), golden #F2B705 accents (pending / due-soon states), burnt orange
#B4552D tertiary (errands, department chart), chocolate #2E2218 headings with taupe body
text, and white cards at ~24px radius with soft shadows — no borders. The palette and fonts
live in tailwind.config.js (cream / sand / mint / mintdark / teal / gold / orange / cocoa /
mocha / taupe) and the blob backdrop in src/index.css. The original live app's slate/blue
system is documented in ../_fleetflow-research/RESEARCH.md.
| Route | Original behaviour reproduced |
|---|---|
/login /register /forgot-password /reset-password | "Welcome back" card, Continue with Google, email+password auth |
/ Missions | Welcome header, Active/Pending stat pills, Today/This Week/This Month/All filters, mission cards, delete with Delete booking for {guest}? confirm, "Pickup in 30 Minutes" notifications |
/mission/:id | The signature flow: camera ODO capture (capture="environment"), simulated AI odometer extraction ("Reading odometer..."), manual fallback offline, Start/End Mission lifecycle, GPS route recording (watchPosition), route trace, remarks, auto distance = end − start |
/fo-dashboard | From/To range, Total Mileage / Active Now / Completed / Pending, "Daily Mileage — Last 14 Days" bar chart, Vehicle Usage chart, Bookings by Type |
/fleet | Drivers / Vehicles / PMS Schedule / Service Logs tabs, Add/Edit/Del modals, per-vehicle Asset Records with full service history, OCR Casa Report capture, Clear All Vehicles |
/fuel | Company fuel expense summary (total spend, liters, avg ₱/L, cost per km, 6-month spend trend, expense share by vehicle) + per-vehicle breakdown (integrity score, consumption trend, flags with evidence, fill history, full-to-full segments) |
/new-booking | Guest Booking or Department Errand mode (Owner, Marketing, Operations Manager, Purchasing, …) — errands run the same ODO + GPS mission workflow; Guest Information / Trip Details / Assignment sections, 6-digit mission_id, FO email notification |
/history | Completed-only, From/To filter, distance totals, Export PDF (print) |
/qr-codes | "Book a Reservation" + "Driver Dashboard" cards via api.qrserver.com, Copy Link |
/settings | Admin-only area (role-gated by login, so it works the same on mobile): User Accounts enrollment with the same quick modal flow as vehicle registration (blank password = auto-generated, Admin/Staff roles, disable instead of delete), sync status, fuel price bands, Danger Zone → "Reset All Transport Data" (keeps Drivers & Vehicles) |
Plus the offline engine: offline mission actions are queued
(create_mileage_log / update_request — the same action shapes as the original),
show "Saved offline. Will sync when back online.", and drain automatically on reconnect,
with a header sync indicator.
These four systems extend the duplicated app (the live Base44 app doesn't have them yet):
1. Fuel consumption & fraud detection (src/lib/fuel.js)
Consumption is computed full-tank-to-full-tank (partials accumulate into the next full
fill) — the industry-honest method. Every fill is audited by seven explainable rules, and
every flag shows its numbers:
| Rule | Severity | Catches |
|---|---|---|
| Tank capacity | red | "Ghost fills" — more liters than the tank holds |
| Odometer sequence | red | Fill logged at a lower odometer than the previous |
| Impossible efficiency | red | km/L above the plausible ceiling (inflated liters / ODO lie) |
| Abnormal thirst | red | km/L below the floor (underfilled receipt, siphoning, leak) |
| Market ₱/L band | amber | Receipt price outside the configurable pump band |
| Statistical outlier | amber | km/L ≥ 2σ from this vehicle's own clean history |
| Unaccounted km | amber | Odometer gap far beyond verified mission km (personal use) |
Each vehicle gets a 0–100 integrity score; the Fuel page opens with a company expense summary (total spend, liters, avg price paid, cost per km, 6-month monthly trend, expense share by vehicle) and drills down to per-vehicle fill history, full-to-full segments, trend chart and flags. Price bands follow DOE/GasWatch PH Metro Manila averages and are editable in Settings.
2. PMS + Renewals & compliance (src/lib/pms.js)
Each vehicle carries a km/month PMS interval — auto-recommended from the model (casa-standard
templates) or manual — due whichever comes first, with the live odometer stream as the km
basis. The Fleet → Renewals board is the one command center for everything that expires:
| Item | Basis | Entry point |
|---|---|---|
| PMS / change oil | km or months, whichever first | interval from model; logs via service |
| Registration (OR/CR) | date (due-soon at 60 days) | OR/CR photo OCR in the vehicle form |
| Insurance | date | policy photo OCR in the vehicle form |
| Tire wear | odometer (due-soon at 85% of life) | tire life + last change; auto-resets when a "Tire Replacement" service is logged |
| Driver's license | date | license card photo OCR in the driver form |
Overdue/due-soon items raise a toast on Fleet load, a ⚠ count on the Renewals tab, status chips on driver and vehicle cards, and stay sorted urgent-first on the board.
3. Document OCR (Fleet → Log Service / vehicle & driver forms)
The same capture → InvokeLLM pipeline handles four documents: the casa/dealer service
report (date, odometer, type, casa, cost, next-service, parts, recommendations → Asset
Record), the driver's license card (name, number, expiry), the LTO OR/CR and the
insurance policy (expiry dates). Extracted fields prefill the forms and the document
photos are kept on the records.
4. Department Errands New Booking has a Guest / Department Errand toggle. Errands (Owner, Marketing, Operations Manager, Purchasing, Finance, HR, …) run the identical workflow — ODO capture, GPS route, mileage, completion — and roll up into a Department Errand Mileage chart on the FO Dashboard for internal chargeback.
5. Roles & user management
Accounts carry a role — Admin (you) or Staff. Settings is the admin's area: the nav
item is hidden from Staff and /settings shows an "Admin access only" lock screen for them,
on any device, because the gate is the login — not the device. Admins enroll the team from
Settings → User Accounts (blank password auto-generates one; accounts can be disabled
instead of deleted; the last active admin can never be deleted). Self-service signups from
the Register page always start as Staff.
6. Cute AI driver avatars (src/lib/avatar.js)
Every driver gets a round, cartoon-style avatar generated from their own license photo —
a genuine client-side stylization pipeline (left-third portrait crop → box-blur smoothing →
6-level posterization → saturation + warm lift → Sobel outlines in theme cocoa). It
auto-generates when a license photo is captured (with a "Regenerate" option), persists on
the driver record, and renders on driver cards with a mint ring; drivers without photos
fall back to initials in a mint circle. Swap generateCuteAvatar for a real image-to-image
endpoint (e.g., an LLM/img2img API) for production-quality cuteness — the wiring stays the same.
| Original (Base44) | Duplicate |
|---|---|
entities.* REST API | localStorage mock in src/lib/db.js with the same list/filter/get/create/update/delete/deleteMany surface |
auth (Google OAuth, email) | local accounts; Google button simulates the OAuth success path |
Core.InvokeLLM ODO extraction | simulated reading after a realistic delay (swap integrations.Core.InvokeLLM in src/lib/db.js for a real endpoint to go live) |
Core.UploadFile | downscaled JPEG data URLs |
Core.SendEmail FO notifications | console log + toast |
| Leaflet/OpenStreetMap route map | SVG polyline trace (zero network deps) |
React 18 · Vite · Tailwind CSS · React Router · lucide-react · Recharts · dayjs — the same library set the original Base44 app compiles to.