Where SSH is enabled for your package, our Web Hosting tools can provide account-level command-line access, including a browser terminal on supported services. This is useful for inspecting application files and logs, running compatible development tools and completing tasks that are easier to express as commands. Available Git tools can also help you keep repositories in the hosting account and organize website code changes.
IN THIS ARTICLE
01 Connect with the correct account and access permissions
Access is limited to the hosting account and the software and permissions provided to it. SSH does not grant root access, control over other customers' files or permission to install arbitrary operating-system services. A command that works on your own computer may require a different executable path, dependency or permission on shared hosting. Check runtime availability before relying on tools such as dependency managers or build commands.
Use the connection details supplied for your own service. If you authenticate with an SSH key, authorize the intended public key through the supported access controls and protect the corresponding private key. Do not paste private keys, passwords or access tokens into ordinary support or AI conversations. Remove unused keys when a developer no longer needs access, and use appropriately limited access arrangements for collaborators.
02 Release and verify application changes
- Confirm the repository, branch, destination folder and the change being released.
- Preserve a suitable backup before changes that could affect a live site.
- Complete the application's required deployment steps and verify the website's important functions.
Pulling code does not necessarily install dependencies, update the database or reload an application. Git history does not contain all business data or uploaded media, so it is not a replacement for the appropriate backup.
Our Hosting AI Engineer can assist with supported development and investigation work inside the account, while the same account permissions and package boundaries continue to apply.
Does a Git checkout recover the application database?
No. A Git checkout changes tracked repository files. It does not restore the live application database, untracked uploads or other account data. Use the appropriate data backup when that information needs recovery, and coordinate it with the deployed code.
- Check the selected repository, branch and destination path before pulling or deploying a change.
- Use supported access and privately stored credentials; verify a changed SSH host identity before trusting it.
- Complete dependency, database and application-startup steps as required, then test the live business function.
Git history records tracked code, not every upload, mailbox, database row or external configuration. Keep the appropriate data backup as well. A browser terminal can offer command and interactive modes on supported accounts, but available access remains within the hosting service permissions and limits.