Are you able to have Dots control your servers/computers?
retained text
I am finding Dots very limited. I have already tried Muse and then Hermes. With both, I can configure them to control my servers and from there do various other jobs. But the Dots agents just seem too limited. Some things I can have my agents do if they were able to have full control of a VM or PC: \- Use the same browsers on my own computers to get around websites blocking the agents. \- Control some apps via MCP servers I am running on my own servers. \- Make changes in Obisidian which helps the agents keep track of details. \- Use other Windows apps. Can't find how to make it connect to my Linux servers. For Windows, it can only control apps I have already opened with some exceptions. If I ask it to control the windows terminal app it gets denied. It simply can't have full control of the server or open new apps. The only workaround I can think of is to create a new Windows app that, itself, has broad access and maybe terminal access. So I approve access to that app and then it has access to my entire PC. Has anyone dealt with this? If I buy a Mac, would THAT have full access?
document fetched 2026-11-08 00:24 · canonical https://reddit.com/r/OpenAI/comments/1xqc6fa/are_you_able_to_have_dots_control_your