signalpilot 0.5.0

SignalPilot CLI - Your Trusted CoPilot for Data Analysis

SignalPilot CLI

Your Trusted CoPilot for Data Analysis - A simple CLI tool to bootstrap Jupyter-powered data science workspaces with AI agent support.

Features

• 🚀 One-command setup - Get from zero to Jupyter Lab in under 3 minutes
• ⚡ Python 3.12 - Uses the latest Python with uv for fast package management
• 📊 Pre-configured workspace - Includes pandas, numpy, matplotlib, seaborn, plotly
• 🤖 AI-ready - Built-in SignalPilot AI agent support
• ⚡ Fast - Optimized Jupyter cache for quick startups
• ✨ Beautiful CLI - Clean, colorful terminal output

Quick Start

# No uv installed: curl -LsSf https://astral.sh/uv/install.sh | sh
uvx signalpilot      # Initialize workspace and starts Jupyter Lab

OR Any time later

# Easy way to start
uvx signalpilot lab  # Start Jupyter Lab in ~/SignalPilotHome

# OR manually activate and start
cd ~/SignalPilotHome && source .venv/bin/activate
jupyter lab

What happens:

• Creates ~/SignalPilotHome workspace with starter notebooks
• Sets up Python 3.12 + Jupyter Lab + data packages (pandas, numpy, matplotlib, plotly)
• Optimizes for fast startup
• Launches Jupyter Lab in your default browser

Why uv?

• 10-100x faster than pip/conda for package installation
• SignalPilot Kernel runs on it - native integration
• Modern Python package management

Other ways to install uv:

# Linux/macOS (recommended)
curl -LsSf https://astral.sh/uv/install.sh | sh

# macOS
brew install uv

What Gets Installed

Python Packages:

• signalpilot-ai - AI agent integration
• jupyterlab - Modern Jupyter interface
• pandas, numpy - Data manipulation
• matplotlib, seaborn, plotly - Visualization
• python-dotenv, tomli - Configuration utilities

Directory Structure:

~/SignalPilotHome/
├── user-skills/       # Custom agent skills
├── user-rules/        # Custom agent rules
├── team-workspace/    # Shared team notebooks
├── demo-project/      # Example notebooks
├── pyproject.toml     # Python project config
├── start-here.ipynb   # Quick start guide
└── .venv/             # Python environment

Advanced Lab Options

Working in Different Directories

# Default: Open ~/SignalPilotHome w/ default .venv
uvx signalpilot lab

# Open in current folder using default .venv in ~/SignalPilotHome
uvx signalpilot lab --here

# Open in current folder using local .venv
# MUST have a .venv in the current directory w/ jupyterlab and signalpilot-ai installed
uvx signalpilot lab --project

--here flag:

• Opens Jupyter Lab in your current directory
• Uses the default environment from ~/SignalPilotHome/.venv
• Perfect for quick exploration in any folder

--project flag:

• Opens Jupyter Lab in your current directory
• Uses a local .venv in that directory
• Great for project-specific work with custom dependencies
• Requires a .venv to exist (create with uv venv --seed --python 3.12)

Passing Extra Arguments to Jupyter Lab

You can pass any Jupyter Lab arguments after the command:

# Custom port
uvx signalpilot lab --port=8888

# Disable browser auto-open
uvx signalpilot lab --no-browser

# Combine with directory flags
uvx signalpilot lab --here --port=8888

# Any jupyter lab flag works
uvx signalpilot lab --ip=0.0.0.0 --port=9999

Requirements

• Python 3.10 or higher
• uv package manager

Permanent Installation Options (Not Recommended)

Option 1: Consider Running with uvx (Recommended - no installation needed)

uvx signalpilot

Option 2: Install with uv

uv tool install signalpilot
sp init

Option 3: Install with pip

pip install signalpilot
sp init

License

MIT License - See LICENSE file for details

Links

• Homepage
• Documentation
• GitHub