/magic:init-workspace

Initialize a MAGIC data processing workspace. Follow these steps:

1. Read the workspace init skill at skills/magic-workspace-init/SKILL.md to understand the workspace directory convention and available options.

2. Determine the workspace path:

   • If the user provided a path argument, use that path
   • Otherwise, use the default: ./workspace/ relative to the current working directory

3. Create the workspace directory structure inline using the SKILL.md patterns (it's safe to run on existing workspaces — never removes files):

   from pathlib import Path
   workspace = Path("<path>")
   for d in ["data/input", "data/checkpoints", "data/output", "logs", "specs", "reports", "charts"]:
       (workspace / d).mkdir(parents=True, exist_ok=True)

4. Verify the Python environment:

   • Check for workspace/.venv/bin/python3 (preferred) or system Python
   • Verify key packages: pandas, numpy
   • Report any missing packages with install guidance

5. Check for existing data files in data/input/ and mention them if present.

6. Present a report with these sections:

   a. Workspace status — directories created or already existed, any existing data files found

   b. Python environment — which Python is available, package status, install guidance if needed

   c. LLM Configuration — check for API keys in environment (GOOGLE_API_KEY, HF_TOKEN, etc.)

   d. Getting started — suggest first steps: load data, run /magic:explore, or /magic:lifecycle

7. Provide a comprehensive overview that helps users understand their workspace setup and next steps.