streamware

🚀 Streamware Quick Start

1-Minute Setup

# Clone
git clone https://github.com/stream-ware/streamware
cd streamware

# Install dependencies
bash install_bot_deps.sh

# Install Streamware
pip install -e .

# Deploy bot
bash deploy_vscode_bot.sh

# Use it!
sq bot continue_work --iterations 5

What Just Happened?

install_bot_deps.sh installed:

• ✅ gnome-screenshot / scrot (for screenshots)
• ✅ Pillow >= 9.2.0 (for image handling)
• ✅ PyAutoGUI (for mouse/keyboard)
• ✅ pyscreeze (for screenshot support)
• ✅ X11 permissions (for display access)

deploy_vscode_bot.sh configured:

• ✅ Ollama + AI models (LLaVA, Qwen2.5)
• ✅ Bot service
• ✅ Control scripts

Now you can:

# Click buttons
sq bot click_button --button accept_all

# Work autonomously
sq bot continue_work --iterations 10

# Take screenshots
sq auto screenshot --text screen.png

# Automate desktop
sq auto click --x 100 --y 200

Troubleshooting

Error: “gnome-screenshot not found”

sudo apt-get install gnome-screenshot scrot

Error: “Can’t connect to display”

xhost +local:
export DISPLAY=:0

Error: “Pillow version too old”

pip install --upgrade "Pillow>=9.2.0"

Run full dependency installer:

bash install_bot_deps.sh

Manual Installation

If automated scripts don’t work:

# 1. System packages
sudo apt-get update
sudo apt-get install -y gnome-screenshot scrot imagemagick xclip xdotool

# 2. Python packages  
pip install --upgrade pip
pip install "Pillow>=9.2.0" pyscreeze pyautogui
pip install SpeechRecognition pyttsx3  # optional: voice
pip install opencv-python              # optional: media

# 3. X11
xhost +local:

# 4. Ollama
curl -fsSL https://ollama.ai/install.sh | sh
ollama pull llava
ollama pull qwen2.5:14b

# 5. Streamware
pip install -e .

Verify Installation

# Test imports
python3 -c "from streamware.components import VSCodeBotComponent; print('✓')"

# Test screenshot
sq auto screenshot --text test.png

# Test bot
sq bot click_button --button accept_all

Next Steps

1. Read the guide: VSCODE_BOT_GUIDE.md
2. Try examples: examples/voice_automation_examples.sh
3. Deploy for real: sq bot continue_work --iterations 100

Support

• 📖 Documentation: docs/
• 🐛 Issues: GitHub Issues
• 💬 Discussions: GitHub Discussions

Your AI pair programmer is ready! 🤖✨

This site is open source. Improve this page.