Vibeleaderboard
Index / app
Visit github.com
Category
Other
Rank
No. 698Tools index
Pricing
Open Source
Type
APP
Latest release
v1.34.0
Added
Apr 14, 2026

About

A self-contained, offline survival computer packed with AI chat, Wikipedia, educational content, maps, and critical tools. Designed to keep you informed and empowered without internet connectivity, perfect for emergency preparedness or remote locations.

What it does

Project NOMAD is a browser-based control plane for assembling and managing local services in Docker containers. Its setup wizard, content managers, app catalog, and update controls turn separate projects for search, learning, maps, notes, data work, and local AI into one deployable system.

Why it's ranked here

The strongest idea is integration, not a novel replacement for every bundled service. NOMAD gives operators one place to configure, update, and extend a substantial offline stack. That convenience carries clear operational costs: Docker knowledge, advance content downloads, careful network controls, and capable hardware for useful local AI.

What's good

The component choices cover distinct jobs instead of duplicating them. Kiwix serves reference material, Kolibri tracks learning, ProtoMaps supplies regional maps, and Qdrant supports document retrieval for chat. Operators can schedule updates, select content, attach a remote model server, install catalog apps, or add custom containers.

Tradeoffs

Installation needs internet access, Debian-based infrastructure, root privileges, and Docker. The management layer is light, but the recommended local AI setup asks for substantial memory, storage, and graphics hardware. Most importantly, NOMAD has no authentication and explicitly relies on network-level access controls. It should not face the public internet without additional security work.

How to use it well

Use it for a household, classroom, workshop, field site, or preparedness server where one operator curates resources before connectivity disappears. Keep it on a controlled local network, download maps and libraries in advance, and test the chosen hardware under realistic load. It does not replace identity management, perimeter security, or administration of a remote AI host.

Technical notes+

The root package.json declares Apache-2.0 but provides only a placeholder test script. admin/adonisrc.ts configures an AdonisJS application with Lucid, Inertia, Edge, sessions, Shield, CORS, static assets, Transmit, and unit plus functional Japa suites. admin/vite.config.ts builds a React and Tailwind client without server-side rendering. admin/bin/server.ts boots the HTTP server and reconciles collection manifests from storage during startup, while admin/ace.js and admin/bin/console.ts expose Ace commands through a TypeScript loader. admin/start/env.ts validates database, Redis, storage, external API, connectivity-test, and creator-pack settings. admin/types/rag.ts models ingestion progress, retrieval results, stored files, and warning states.

Observed

License
Apache-2.0
Primary language
TypeScript, with JavaScript entrypoints and configuration files
Install surface
Root-required terminal installer for Debian-based systems, with a manual Docker Compose option
Interfaces
Browser management UI, HTTP API, and Ace command-line commands
Packaging
Docker-based orchestration of the management application and optional services
Platform support
Debian-based operating systems; Windows through a community-supported WSL2 path
Frontend structure
React client built with Vite, Inertia, and Tailwind CSS
Test structure
AdonisJS configuration defines unit and functional Japa test suites, while the root package test script is a failing placeholder

Read from README.md, package.json, admin/ace.js, admin/adonisrc.ts, admin/vite.config.ts, admin/eslint.config.js, admin/tailwind.config.ts, admin/bin/test.ts, admin/start/env.ts, admin/types/rag.ts, admin/bin/server.ts, admin/config/app.ts, admin/types/chat.ts, admin/types/maps.ts, admin/bin/console.ts.

What it can do

  • Answer questions using AI chat with retrieval-augmented generation

    Natural language questionsAI-generated responses with relevant context from knowledge base

  • Search and browse Wikipedia articles offline

    Search terms or article titlesWikipedia article content and related pages

  • Access Khan Academy educational courses and lessons

    Course or subject selectionEducational video content and learning materials

  • Display and navigate offline maps

    Geographic coordinates or location namesInteractive map displays with navigation capabilities

  • Perform data analysis on uploaded datasets

    Data files or datasetsStatistical analysis results and visualizations

  • Manage and orchestrate containerized applications

    Container configurations and commandsRunning containerized services and status reports

Tags

offlinesurvivalaieducationmapsemergencyself-hosteddocker

Tech Stack

Node.jsDocker

Featured in Intel

Comments (0)

No comments yet

Indexed by a proprietary survey. Corrections welcome.