Workspaces

Workspaces are project directories that you connect to the RenX platform. When you start a session, your agent works within a workspace — reading files, writing code, running commands, and managing project resources.

Who this is for: Anyone who wants an agent to work on local or remote project folders.

Connecting a Workspace

When you start a new session, you select a workspace folder:

1. From the Start Session card, click the Folder dropdown.
2. Choose from:
   • Known workspaces — Directories from connected machines.
   • Recent directories — Folders you’ve used before.
   • Browse… — Open a file picker to select any directory.

The selected folder becomes the working directory for your agent’s session.

Multi-Machine Support

RenX supports working across multiple machines. If you have RenX running on another computer, it will appear in the Machine dropdown when starting a session.

This allows you to:

• Start a session on your laptop that works on files on your desktop.
• Run commands on a remote machine from your local RenX instance.

Each machine is identified by a name and unique ID, visible in the Workspaces tab.

Workspaces Tab

Click the Workspaces tab (monitor icon) in the sidebar to see all your workspaces:

• Online workspaces — Currently active and connected.
• Offline workspaces — Previously connected but not currently running.

Each workspace shows:

• Path (e.g., ~/Code/my-project)
• Machine name
• Platform and architecture
• Git repository status
• Connection status (online, reconnecting, offline)

Workspace Configuration

Select a workspace to configure its settings:

Skills

View and manage the automation skills available in this workspace.

MCP Servers

See which Model Context Protocol servers are configured for this workspace. See MCP for details.

Permissions

Configure what your agent is allowed to do in this workspace:

• Allow list — Operations the agent can perform without asking.
• Deny list — Operations the agent is never allowed to perform.
• Ask list — Operations that require your approval each time.

Hooks

Set up pre- and post-execution hooks that run before or after agent actions.

Plugins

View installed plugins and their status for this workspace.

File Operations

When an agent is working in a workspace, it can:

• Read files — View source code, configuration files, documentation.
• Write files — Create new files or modify existing ones.
• Upload files — Add files to the workspace.
• Delete files — Remove files (with permission).

All file operations are subject to the permission controls you’ve configured. Sensitive operations will trigger permission requests.

Terminal Sessions

Your agent can create terminal sessions within a workspace to execute commands. This is useful for:

• Running build tools and test suites.
• Installing dependencies.
• Interacting with version control (git).
• Running development servers.

Terminal commands execute on the machine where the workspace is located, not necessarily on your local machine.

Local-First Storage

Workspace settings, project files, and cached data are stored locally on your device. They are not uploaded to Open Mercury’s servers unless needed for a specific feature (such as multi-device session continuity).

When you disconnect a workspace or close the app, your local files remain untouched.

Next Step

• Return to Getting Started and launch a session from the Start Session card.
• Go to Built-in agents if you want to assign a different agent personality or model setup to the workspace.
• Go to Extensions if you want workspace-scoped skills, plugins, hooks, or permissions.