Sider Omni Sidebar - Give every Mac app an Agent Sidebar

An AI agent beside any Mac app. Sider Omni sidebar sees what's on your screen and acts on it directly, right where you're working. Watch the result, ask for the next change. Nothing to copy, nothing to upload. Built on GPT-6 Astra, and it can run on your existing ChatGPT or Codex credits.

Add a comment

Replies

Best

I wonder how well it handles workflows across apps like taking something from Safari and using it in Notes or another tool.

 You can definitely give it a try first.

Maker

 Cross-app tasks like this are one of Sider Omni’s main use cases. It can organize content from Safari and add it to Notes or another app, saving you from copying and pasting back and forth.

Does Omni support keyboard shortcuts for quickly opening the sidebar or sending the current task?

 Yes, you can press both the left and right Option keys on the keyboard at the same time to turn it on or off, and this will be customizable in the future.

Maker

 Yes. By default, pressing the left and right Option keys together opens Sider Omni, and you can change the shortcut in Settings. Once you’ve typed a task, press Enter to send it.

​can it interact with non native web apps inside browser windows or just desktop apps?

Maker

 It works across native desktop apps, Safari, browser-based web apps, and even apps with custom-drawn interfaces. Chrome also has a dedicated Sider Operator extension for more powerful browser use.

 The browser is an app too, so of course it works there as well, but it can't operate the page as efficiently as our extension product does—it can only work through computer use.

​am seeing what is on screen without uploading files is definitely how all desktop ai tools should work.

Maker

 That was a big reason we built Sider Omni. It stays with the app you’re using, understands what’s in the window when you need help, and can get the task done without making you upload files or copy and paste everything first.

 Yes, you can give it a try.

​I am just downloaded it to test with VS Code! curious to see how well it picks up code context without manual prompts.

Maker

 Awesome—would love to hear how it goes! When you start a task in VS Code, Sider Omni can understand the code context in the current window without making you copy it into the chat first.

 Thank you, Adams, I would really appreciate it if you could share your feedback with me.

The no copy and no upload workflow is an interesting approach. Does it keep any screen content locally when working with sensitive apps?

 There will be a cache, and you can clear it periodically. A clear cache button will be added in settings later.

Maker

 When a task requires screenshots, Sider Omni may temporarily keep the relevant captures locally to complete and verify the task. It doesn’t continuously record or save your screen in the background.

Congratulations on the launch! Having the bot by the side of the application rather than in another separate chat window is the right approach in my opinion.

Maker

 Thank you! We wanted you to be able to ask for help or get something done without stepping away from what you’re working on.

The Numbers chart example made sense to me. I spend half my time shifting information between applications so having a chart improved through asking what comes next seems very appealing.

Maker

 Thanks! Charts are rarely done in one pass—you see the result, then want to tweak the data or how it’s presented. Being able to keep that conversation going beside Numbers saves a lot of back-and-forth. If you run into any issues, please let us know. I really want to get this right.

I wonder how it works in applications that have custom UIs, for example, Figma or Logic Pro. Does it depend on the accessibility labels in macOS or does it use mostly screen shots?

Maker

 It uses both, depending on what the app exposes. When macOS accessibility information is available, Sider Omni uses it to identify text and controls. For custom-drawn interfaces, it can also use screenshots to understand and interact with what’s visible. So it doesn’t rely solely on accessibility labels, though we’re still refining the experience across different apps.

As an avid user of client emails via Mail, my one bad practice is to retype the response three times. It will save me a few minutes if Sider Omni could read the entire email and type right from there. Does it support attachment as well or just visible text?

Maker

 Yes—you can ask Sider Omni to read the email and help draft or revise your reply right in Mail. If an attachment matters to the response, you can ask it to open that too; what it can read depends on the file type.