The Last Caretaker - Human Optimizer

A static web app for planning and optimizing food + memory builds for professions in The Last Caretaker, including inventory-aware planning, progress tracking, and a mobile second-screen share flow.

Warning

Spoilers ahead: this project includes profession, trait, planning, and progression information that may reveal discovery content in The Last Caretaker.

What This App Does

The app helps you:

• Optimize food and memory combinations for a selected profession.
• Track inventory and calculate deficits before crafting.
• Build multi-session human plans with rocket grouping.
• Track created/planned profession progress by committee.
• Share a read-only mobile snapshot of your plan via QR code.

No backend is required for runtime. The app is fully client-side and hosted as a static site.

Core Workflows

Profession optimization

Use the Optimizer tab to select a profession, set baseline physical floors, apply exclusions/toggles, and generate a recommended plan.

• Physical traits (Life Exp, Height, Weight, Strength, Intellect) are satisfied by food.
• Personality traits are satisfied by memories.
• Hard minimum floors: Life Exp >= 10, Height >= 30, Weight >= 20.

Planning and deficit review

Use the Plan tab to accumulate humans into sessions and rockets.

• Sessions support up to 4 planned humans.
• Rockets support up to 3 humans.
• Deficit tables show what inventory is missing globally and per session.

Second-screen mobile companion (QR share)

If you play fullscreen and want the plan visible on phone:

1. Build/update your plan on PC.
2. In Plan, click Share plan via QR.
3. Scan with your phone.
4. Open the read-only mobile page (share.html) as your companion view while gaming.

Notes:

• Share payload is compressed into URL hash (no server storage).
• You can share all sessions or selected sessions only.
• QR size guidance is shown in-app (balanced and ultra-safe subset suggestions).
• Shared page defaults to the sharer's theme, while viewers can still switch theme on mobile.
• Data tab can generate QR snapshots for inventory-only or created-professions progress-only sharing.

App Feature Map

• Home: overview, workflow guidance, and project notice.
• Optimizer: profession-targeted plan generation with inventory-aware options.
• Sandbox: free-form quantity testing and resulting profession fulfillment.
• Plan: sessions, rockets, deficits, and QR/share generation.
• Progress: created/planned/remaining tracking by committee.
• Professions: browsable requirements and status marking.
• Foods: food reference and inventory editing.
• Memories: memory reference and inventory editing.
• Map: filterable map locations reference with pan/zoom on the 100m grid.
• Data: import/export profile JSON and reset inventory.
• Data: import/export profile JSON, reset inventory, and generate inventory/progress QR snapshots.

Architecture Overview

Runtime architecture (app first)

This project is a browser-only static app with two entry pages:

• index.html: main planner SPA (tab-based UI).
• share.html: mobile-focused read-only plan viewer for shared links.

JavaScript is split by responsibility:

• js/optimizer.js: optimization engine and domain logic.
• js/ui.js: UI rendering, state orchestration, persistence, plan/share actions.
• js/share.js: decode/render logic for shared mobile snapshots.

Styling is split by page:

• css/styles.css: main app styles and themes.
• css/share.css: mobile share page styles and themes.

Domain data is pre-generated/static:

• data/foods.js
• data/memories.js
• data/professions.js
• data/bio.js (manually maintained source metadata and conversion helpers)

Persistence model

Runtime data is stored in browser localStorage (per browser profile/device), including:

• Inventory (foods/memories)
• Sandbox values + filters
• Plan sessions/humans/flights
• Created professions
• Selected tab/mode/theme and optimizer settings

Import/export JSON is provided in the Data tab for portability.

Share model

Sharing uses compressed URL hash payloads:

• No backend persistence
• Read-only mobile rendering by default
• Optional import into local planner state from the shared snapshot
• Supports typed snapshots:
  • plan (sessions/flights/humans)
  • inventory (current food + memory quantities)
  • progress (created professions)

Running Locally

Open index.html directly in a browser.

Optional local server options:

• VS Code + Live Server extension
• python -m http.server 8080 then open http://localhost:8080
• npx serve .

Hosting and Deployment

Production is hosted on GitHub Pages:

• https://camrex.github.io/last-caretaker-human-optimizer/

For setup and release checklist, see:

• docs/deploy.md

Key deployment notes:

• Push to main triggers Pages rebuild.
• App runtime remains fully static (no server runtime).

Data Pipeline (Maintainer "Backend")

Although the app runtime is static, data maintenance uses a Python sync pipeline.

Canonical source workbook:

• resource/The Last Caretaker - Human Needs.xlsx

Sync script:

• scripts/sync_tables_from_excel.py

Commands:

• Write generated data:
  • e:/DevProjects/last-caretaker-human-optimizer/.venv/Scripts/python.exe scripts/sync_tables_from_excel.py --write
• Check whether generated files are in sync:
  • e:/DevProjects/last-caretaker-human-optimizer/.venv/Scripts/python.exe scripts/sync_tables_from_excel.py --check

Pipeline output:

• Regenerates data/foods.js, data/memories.js, data/professions.js in sparse format (non-zero numeric fields only).
• Emits stable numeric IDs for foods/memories/professions and persists name->id mapping in resource/item_ids.json for compact share encoding.
• Leaves data/bio.js manually maintained.

Repository Structure

last-caretaker-human-optimizer/
├── index.html
├── share.html
├── css/
│   ├── styles.css
│   └── share.css
├── js/
│   ├── optimizer.js
│   ├── ui.js
│   └── share.js
├── data/
│   ├── bio.js
│   ├── foods.js
│   ├── memories.js
│   └── professions.js
├── scripts/
│   └── sync_tables_from_excel.py
├── resource/
│   └── The Last Caretaker - Human Needs.xlsx
└── docs/
        └── deploy.md

Privacy and Constraints

• All runtime state is local to the browser/device unless explicitly exported/shared.
• Shared links can expose planning details to anyone with the URL.
• QR scan reliability depends on payload size; prefer session-only sharing when needed.

Planned / TBD

• Human tracking: track each created human with created date, profession, committee assignment, and Earth/Space status (allow duplicate professions across multiple humans).
• Session/Rocket flow update: allow rocket planning without requiring humans to be marked as created first.
• Human tracking + rocket planning bridge: support adding already-created humans (outside a plan) into rocket plans.
• Human seed tracking: track seed availability and remaining seeds per known location.
• Ensure full data portability: include all plan/tracking state in Data Export/Import.
• Add a non-spoiler mode that hides or masks progression-sensitive data until explicitly revealed.