self-contained-python-script

You can run a Python script with dependencies without a virtual environment by using PEP 723 inline metadata to declare packages and executing it with uv. This automatically installs the required libraries in an isolated environment at runtime.

PEP 723 inline metadata is a standard for declaring dependencies directly within a Python script file. It eliminates manual environment setup by allowing tools like uv to read the script header and automatically install required packages.

To create a Python script that automatically installs packages using uv, add the provided uv shebang and PEP 723 inline metadata to your file. This configures the script to run in an isolated environment with the declared dependencies.

Yes, you can share a single Python file requiring external libraries like httpx by using PEP 23 inline metadata. The script becomes self-contained and reproducible, executing correctly on other systems without requiring manual pip installs.

Yes, uv supports executing standalone Python CLIs with declared dependencies. It reads the PEP 723 metadata embedded in the script to manage the isolated environment and automatically satisfy the required packages.

PEP 723 is limited to single-file Python scripts and does not replace full project environments. It is best suited for standalone tools and CLIs managed by uv, rather than complex applications requiring multi-file dependency tracking.