Native SSH Remote Projects and Remote Agent Execution

Discussion1

• Rafael Kress1 month ago

  When working with a project located on a remote machine, Agentroom currently requires the project location and agent execution environment to be configured separately. This creates unnecessary configuration overhead. A remote project may be accessible through a mounted filesystem while agents still execute locally unless they are explicitly configured otherwise. For projects opened through a native remote connection, Agentroom should automatically use the remote host as the default execution environment for agents associated with that project. Ideally: - Opening a remote project automatically selects the corresponding remote host for agent execution. - Terminals and project commands use the same remote environment. - Users can override this behavior if they intentionally want an agent to run somewhere else. - The execution environment should be visible in the UI so that it is always clear whether an agent is running locally or remotely. This would make remote projects behave as a coherent development environment instead of requiring separate configuration for file access and agent execution.

You will be told by email and in the app on every reply. Every message carries a link to stop them.