Interactive setup wizard for Garuda System — entity-aware web intelligence platform (Closed Alpha)
Project description
Garuda Setup
⚠️ Closed Alpha — Garuda Intelligence is currently in closed alpha. Access requires registration at garuda.nxs.solutions and a Gitea account on git.nxs.solutions. Docker image pulls require authentication against
docker.nxs.solutions.
Interactive cross-platform setup wizard for Garuda System — the all-in-one intelligence platform by Garuda Intelligence.
Creates a self-contained install directory with docker-compose.yml,
.env, data directories, and platform-native launcher scripts.
No source checkout needed — just install, run the wizard, and go.
About Garuda
Garuda is an advanced, entity-focused intelligence gathering and analysis platform that combines web crawling, LLM-powered extraction, semantic search, and adaptive learning to build comprehensive knowledge graphs. Named after the mythical eagle with omniscient vision, Garuda provides deep insights into entities — companies, people, organizations, products — through intelligent, gap-aware data collection.
Key Capabilities
| Feature | Description | |
|---|---|---|
| 🧠 | Intelligent Crawling | Analyzes what it already knows and automatically fills knowledge gaps through targeted searches |
| 🔍 | RAG-First Hybrid Search | SQL keyword + semantic vector search with automatic retry using paraphrased queries (4-phase search) |
| 🕸️ | Knowledge Graph | Tracks entities, relationships, and temporal changes with full provenance |
| 🎬 | Multi-Modal Processing | Extracts intelligence from text, images (OCR), video (transcription), and audio |
| 📁 | Local Data Ingestion | Upload files or watch directories — PDF, CSV, JSON, media — processed through the same extraction pipeline |
| 🗄️ | Multi-Database | Create isolated databases per domain, instant switching, merging, and cross-database global search |
| 📊 | Adaptive Learning | Learns from crawl results to improve future discovery strategies |
| 🔧 | Production-Ready UI | Web UI with 16 specialized panels, REST API, and Chrome extension |
Built-in Components
The Garuda System container ships everything in a single image:
- Web UI — 16-panel Flask interface (Agent, Chat, Crawl, Entities Graph, Intel, Search, Media, and more)
- IDE & REPL — Browser-based development environment with GLang (Garuda-Lang) scripting
- Ollama — Local LLM inference (llama3.2, granite, mistral, and others)
- Qdrant — Vector database for semantic/embedding search
- PostgreSQL — Structured data storage
- Chrome Extension — Record pages and elements directly from your browser
- CLI Tools —
garuda,garuda-search,garuda-agent,garuda-db,garuda-explorer, and more
Use Cases
- 🎯 OSINT — Gather structured intelligence on companies, individuals, infrastructure
- 🏢 Corporate Research — Entity profiles with automated gap analysis and competitive intelligence
- 🚨 Threat Intelligence — Track actors, infrastructure, and tactics from public sources
- 📚 Academic Research — Build domain-specific knowledge graphs from scholarly and news sources
- 📡 Brand Monitoring — Monitor mentions, sentiment, and relationships across the web
Tech Stack
Python 3.10+ · Flask · SQLAlchemy · Qdrant · Ollama · Selenium · Tesseract OCR · sentence-transformers · Docker · Typer/Rich CLI
For full documentation, architecture diagrams, and API reference, see the main repository on Gitea.
Installation
From PyPI (recommended)
pip install garuda-setup
From Garuda Package Registry
pip install garuda-setup \
--index-url https://pypi.nxs.solutions/api/packages/garuda/pypi/simple
From Source
git clone https://git.nxs.solutions/garuda/Garuda.git
cd Garuda/setup
pip install .
Run the Wizard
garuda-setup
Or without installing:
python -m garuda_setup
Platform Support
| Platform | Docker Runtime | GPU Passthrough | Launcher |
|---|---|---|---|
| Linux | Docker Engine | ✅ NVIDIA (via nvidia-container-toolkit) | garuda.sh |
| macOS | Docker Desktop / Colima | ❌ (CPU-only; use Garuda Cloud for GPU) | Garuda.command |
| Windows | Docker Desktop (WSL 2) | ✅ NVIDIA (via WSL 2) | garuda.bat / garuda.ps1 |
macOS Prerequisites
# Docker Desktop (recommended)
brew install --cask docker
# Or Colima (lightweight alternative)
brew install colima && colima start
Apple Silicon (M1–M4): Fully supported. The container image is multi-arch. GPU passthrough is not available in Docker on macOS — connect a remote GPU via Garuda Cloud instead.
What It Does
The wizard walks you through a guided setup and generates everything needed to run Garuda System on your machine:
- Detects your environment — OS, Docker runtime, GPU availability
- Asks for configuration — ports, services, credentials, mount paths
- Creates the install directory:
~/garuda/
├── docker-compose.yml # pulls from docker.nxs.solutions
├── .env # all configuration
├── garuda.sh # launcher (Linux)
├── Garuda.command # launcher (macOS — double-click in Finder)
├── garuda.ps1 # launcher (Windows PowerShell)
├── garuda.bat # launcher (Windows CMD)
├── README.md # quick reference
├── data/
│ ├── garuda/ # databases, indexes, projects
│ ├── qdrant/ # vector search storage
│ ├── postgres/ # PostgreSQL data
│ └── ollama/ # downloaded LLM models
├── home/ # persistent container home
└── watch/ # drop files here for auto-ingestion
- Optionally pulls the image from
docker.nxs.solutions/garuda/garuda-system:latestand starts the container immediately
Configuration Options
| Option | Description | Default |
|---|---|---|
| Install directory | Where all data and config live | ~/garuda |
| API key | Garuda UI access key (auto-generate, custom, or skip) | auto |
| Port | Web UI port | 7331 |
| IDE | Enable the built-in code/intelligence IDE | yes |
| Ollama | Built-in LLM inference engine | yes |
| GPU passthrough | NVIDIA GPU access (Linux/WSL only) | auto-detect |
| Shell access | SSH into the container | no |
| VNC desktop | Remote desktop via browser | no |
| External PostgreSQL | Connect to an existing database | — |
| External Qdrant | Connect to an existing vector DB | — |
| External Ollama | Use a remote Ollama instance | — |
| AI models | LLM and embedding model selection | llama3.2 / nomic-embed-text |
| Cloud | Connect to Garuda Cloud (GPU provisioning, sync) | — |
| Credential mounts | GitHub Copilot config, GH CLI tokens, SSH keys | auto-detect |
| Host mounts | Additional directories to observe/index | — |
Launcher Commands
After setup, use the generated launcher:
cd ~/garuda
./garuda.sh up # start all services
./garuda.sh down # stop all services
./garuda.sh logs # follow container logs
./garuda.sh update # pull latest image & restart
./garuda.sh status # show container status
./garuda.sh shell # open a bash shell in the container
./garuda.sh config # display current configuration
Windows: use garuda.bat or garuda.ps1 with the same sub-commands.
macOS: double-click Garuda.command in Finder to start and open your browser,
or use ./garuda.sh from Terminal.
Docker Image
The container image is hosted on the Garuda private Docker registry:
docker.nxs.solutions/garuda/garuda-system:latest
During the closed alpha, pulling the image requires authentication with your git.nxs.solutions credentials:
docker login docker.nxs.solutions
# Username: your git.nxs.solutions username
# Password: your git.nxs.solutions password
The setup wizard handles image pulling automatically if Docker is available and you are already logged in.
Getting Access (Closed Alpha)
Garuda Intelligence is in closed alpha. To get started:
- Register on git.nxs.solutions — this gives you access to the source code, package registry, and Docker registry
- Register on garuda.nxs.solutions — this gives you access to Garuda Cloud (remote GPU provisioning, sync, team features)
- Login to the Docker registry:
docker login docker.nxs.solutions
- Install and run the setup wizard:
pip install garuda-setup garuda-setup
If you have trouble accessing any of the services, reach out via the issue tracker.
Links
| Resource | URL |
|---|---|
| Garuda Intelligence (main site) | garuda.nxs.solutions |
| Source Code (Gitea) | git.nxs.solutions/garuda/Garuda |
| Package Registry (Gitea PyPI) | pypi.nxs.solutions |
| Docker Registry | docker.nxs.solutions |
| PyPI (public mirror) | pypi.org/project/garuda-setup |
| Issue Tracker | git.nxs.solutions/garuda/Garuda/issues |
License
MIT — see LICENSE for details.
Garuda Intelligence — Closed Alpha
garuda.nxs.solutions ·
git.nxs.solutions ·
docker.nxs.solutions ·
pypi.nxs.solutions
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file garuda_setup-1.0.2.tar.gz.
File metadata
- Download URL: garuda_setup-1.0.2.tar.gz
- Upload date:
- Size: 19.7 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
5580632991a248fc57ecdf6b75207d61f0e81d831a0cb2026eaec11febd46186
|
|
| MD5 |
28b9f9a080589d17bab201948062f46f
|
|
| BLAKE2b-256 |
c91478ff9360da9e8eed33165982ec99fdb1f0d2fd1de7704f18184e2f2d0958
|
File details
Details for the file garuda_setup-1.0.2-py3-none-any.whl.
File metadata
- Download URL: garuda_setup-1.0.2-py3-none-any.whl
- Upload date:
- Size: 18.5 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
b84607c0b71fd716e82fd362c91df0f5ed368193dab4728cb90cf5f229e58c11
|
|
| MD5 |
45a322732b3e7153363821a1b196cf37
|
|
| BLAKE2b-256 |
0a2487dd0a17e3cd551d268a3f745299f0b72fb516e74989e955898e4ebb2733
|