GNCore is published as a Python package. Install the CLI on Windows, macOS, or Linux, then choose the exact skills your assistant needs. No website account required.
These commands use a Python virtual environment for isolation, avoiding system Python conflicts. Choose the tab matching your operating system and paste the steps into a terminal.
Windows 10 / 11 · PowerShell — install Python 3.10+ first if py --version fails. These commands run without activating a PowerShell script.
py --version
py -m venv .venv
.\.venv\Scripts\python.exe -m pip install --upgrade pip gncore
.\.venv\Scripts\gncore.exe version
.\.venv\Scripts\gncore.exe list appsIn PowerShell, use .\.venv\Scripts\gncore.exe while the virtual environment is not activated.
macOS · Terminal (zsh/bash) — ensure Python 3.10+ is available. A virtual environment avoids externally managed Python installation issues.
python3 --version
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip gncore
gncore version
gncore list appsRun source .venv/bin/activate in later terminal sessions before using the gncore command.
Linux / Windows Subsystem for Linux — ensure python3 and its venv support are installed using your distribution package manager.
python3 --version
python3 -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip gncore
gncore version
gncore list appsFor Ubuntu / Debian, install the Python venv package matching your Python version if python3 -m venv is unavailable.
Install results depend on the available PyPI release and your local Python environment. The GNCore GitHub source currently identifies version 4.0.0; verify the installed version with gncore version.
Find a skill by name or job, install by real catalog ID into a supported target, and validate the result. Workspace scope keeps the installation specific to your current project.
# When using the no-activation PowerShell instructions above:
.\.venv\Scripts\gncore.exe search "clean data"
.\.venv\Scripts\gncore.exe install sheet-clean --apps vscode-codex --scope workspace
.\.venv\Scripts\gncore.exe validate --apps vscode-codexgncore search "clean data"
gncore install sheet-clean --apps vscode-codex --scope workspace
gncore validate --apps vscode-codexWhen you already have an appropriate managed Python environment, install with pip.
python -m pip install --upgrade gncoreOn macOS and Linux, python3 -m pip may be appropriate. Prefer a virtual environment to avoid modifying system packages.
If pipx is installed, keep GNCore isolated while placing its executable on your PATH.
pipx install gncore
pipx upgrade gncoreSet up pipx first using its official instructions and check that your shell can find the GNCore executable.
For developers modifying GNCore, install from the public GitHub repository rather than using a published wheel.
git clone https://github.com/paladuguganeshnaidu/gncore.git
cd gncore
python -m pip install -e .
gncore versionRun inside your own Python 3.10+ virtual environment.
Create a ZIP per skill and manually upload it through the supported ChatGPT Skills interface. This is not a direct ChatGPT integration or automatic sync.
gncore export --app chatgpt --skills company-ceo --output company-ceo.zipCheck ChatGPT account eligibility and workspace upload requirements before using.
GNCore provides health checks and managed backup/restore commands. The instructions guide an assistant, but they do not independently execute an AI model.
gncore doctor
gncore list apps
gncore validate --apps vscode-codex
gncore backup --output gncore-backup.zip
gncore update --allThe package's currently installed version determines available command flags. Review the CLI usage documentation ↗ for details.