Tara Menon c78a8e82b5 Initial release: GyanDoot v2.0.0
Cross-platform AI Agent Platform
- Sources: https://git.gyannetra.in/gnpl_admin/gyandoot-v2
- Installer: curl -fsSL https://git.gyannetra.in/gnpl_admin/gyandoot-release/raw/main/install.sh | bash

Platforms:
  - macOS (arm64/x86_64) via Homebrew or install.sh
  - Linux (x86_64/aarch64/i686) via install.sh
  - Windows (x86_64) via Git Bash/MSYS2 or PowerShell
  - WSL via install.sh

Includes:
  - install.sh (cross-platform universal installer)
  - Windows: install.bat + install.ps1
  - macOS: Homebrew formula (homebrew/gyandoot.rb)
  - Linux: prebuilt binary + install guide
  - README.md with full documentation
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30
2026-06-13 18:33:38 +05:30

🚀 GyanDoot v2.0 — AI Agent Platform

GyanDoot (ज्ञानदूत) — "Knowledge Messenger" — your AI agent that searches, browses, and executes tasks autonomously.


📦 Quick Install

🐧 Linux / 🍎 macOS / WSL

curl -fsSL https://git.gyannetra.in/gnpl_admin/gyandoot-release/raw/main/install.sh | bash

🪟 Windows (PowerShell)

iwr -useb https://git.gyannetra.in/gnpl_admin/gyandoot-release/raw/main/windows/install.ps1 | iex

🍺 macOS (Homebrew)

brew tap gnpl_admin/gyandoot
brew install gyandoot

🖥️ Platform Support

OS Architecture Status
macOS arm64 (M1/M2/M3/M4), x86_64 ✅
Linux x86_64, aarch64, i686 ✅
WSL (Windows) x86_64, aarch64 ✅
Windows (native) x86_64 ✅ (Git Bash / MSYS2)

📋 Prerequisites

Dependency Why Install
Rust (cargo ≥ 1.70) Build GyanDoot binary `curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs
Node.js (≥ 20) Camoufox browser daemon nvm install 22 or nodejs.org
Linux deps Camoufox browser engine See Linux Dependencies section

🔧 Manual Installation

Step 1: Clone & Build

git clone https://git.gyannetra.in/gnpl_admin/gyandoot-release.git
cd gyandoot-release
./install.sh

Step 2: Verify Installation

gyandoot --help
gyandoot setup    # Interactive config wizard
gyandoot run      # Start the agent

Step 3: Browser Setup (Anti-Detection)

# Download Camoufox browser binary (~311MB)
npx --yes camoufox-js@0.11.0 fetch

# macOS only — remove Gatekeeper quarantine
xattr -dr com.apple.quarantine ~/Library/Caches/camoufox

# Verify
ls ~/Library/Caches/camoufox/version.json

🐧 Linux Dependencies

Camoufox (Playwright-based) needs system libraries. Install with:

# Debian / Ubuntu / WSL
sudo apt-get update && sudo apt-get install -y \
  libgtk-3-0 libnss3 libnspr4 libatk1.0-0t64 \
  libatk-bridge2.0-0t64 libcups2t64 libdrm2 \
  libxkbcommon0 libxcomposite1 libxdamage1 \
  libxfixes3 libxrandr2 libgbm1 libpango-1.0-0 \
  libcairo2 libasound2t64 libxcb-shm0

# Fedora / RHEL
sudo dnf install -y nss nspr atk at-spi2-atk cups-libs \
  libdrm libXcomposite libXdamage libXrandr \
  mesa-libgbm pango alsa-lib libxkbcommon

🪟 Windows (Native)

Using Git Bash / MSYS2

# Install Rust
curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh

# Install Node.js
# Download from: https://nodejs.org (v20+)

# Build & install
git clone https://git.gyannetra.in/gnpl_admin/gyandoot-release.git
cd gyandoot-release
./install.sh

Using PowerShell

# Run as Administrator
Set-ExecutionPolicy Bypass -Scope Process -Force
iex ((New-Object System.Net.WebClient).DownloadString('https://git.gyannetra.in/gnpl_admin/gyandoot-release/raw/main/windows/install.ps1'))

🍺 macOS Homebrew

# Tap the repository
brew tap gnpl_admin/gyandoot

# Install
brew install gyandoot

# Upgrade
brew upgrade gyandoot

🚦 Getting Started

# 1. Configure your LLM provider
gyandoot setup

# 2. Start interactive session
gyandoot run

# 3. Browser-based tasks (with anti-detection)
gyandoot browser search "latest AI research papers"
gyandoot browser browse https://example.com

# 4. Service management (auto-start on boot)
gyandoot service install   # optional
gyandoot service status

🏗️ Architecture

gyandoot run
  ├── Agent Loop (LLM-powered)
  ├── Browser Manager (Camoufox daemon)
  │   ├── camoufox-js@0.11.0  → fetches browser binary
  │   └── camoufox-browser@0.1.0 → daemon IPC (Unix socket)
  ├── Memory Palace (Persistent recall)
  │   └── Vector-search memory at ~/.gyandoot/palace/
  └── Tool Executor
      ├── web_search / web_browse (Camoufox browser)
      ├── file_ops, code_exec
      └── subagent delegation

📜 License

MIT — Gyannetra Pvt Ltd

Description
GyanDoot v2.0 — Cross-platform AI Agent Platform (release builds + installers)
Readme 4.7 MiB
Languages
Shell 59.2%
PowerShell 20.5%
Batchfile 16.1%
Ruby 4.2%