Installation
How to install skene using uvx, pip, or from source.
Prerequisites
- Python 3.11 or later. Check your version with
python3 --version.
Option 1: uvx (Recommended)
uvx runs Python CLI tools without installing them globally. This is the fastest way to start using skene.
If you don't have uv installed yet:
curl -LsSf https://astral.sh/uv/install.sh | sh
Then run any skene command directly:
uvx skene analyze .
uvx skene plan
No pip install needed. uvx downloads the package into an isolated environment on first run and caches it for subsequent calls.
Option 2: pip
Install skene into your current Python environment:
pip install skene
After installation, the skene and skene commands are available on your PATH:
skene analyze .
skene chat
Option 3: From source
Clone the repository and install in development mode:
git clone https://github.com/SkeneTechnologies/skene.git
cd skene
Using uv (recommended):
uv sync
Using pip:
pip install -e .
To include all optional dependencies:
pip install -e ".[mcp,ui]"
Optional extras
Install extras for additional functionality:
| Extra | What it adds |
|---|---|
mcp | MCP server support (skene-mcp entry point). Adds mcp>=1.0.0 and xxhash>=3.0. |
ui | Interactive terminal prompts via questionary>=2.0. |
# With uvx (use --from to specify extras)
uvx --from "skene[mcp]" skene-mcp
# With uv
uv pip install skene[mcp]
# With pip
pip install skene[mcp]
pip install skene[mcp,ui]
Verifying the installation
Run the version check to confirm everything is working:
skene --version
Expected output:
skene 0.2.1
If you used uvx:
uvx skene --version
Next steps
Proceed to the Quickstart to configure an LLM provider and run your first codebase analysis.