SidekickBar

SidekickBar

Your new multi-agent AI sidekick

140 followers

Boost your productivity with 30+ AI assistants like ChatGPT, Copilot, Claude & Gemini in one sidebar. Write, code, translate & create faster with AI across Windows, macOS, and Linux.
SidekickBar gallery image
SidekickBar gallery image
SidekickBar gallery image
Free Options
Launch Team / Built With
Intercom
Intercom
Startups get 90% off Intercom + 1 year of Fin AI Agent free
Promoted

What do you think? …

Manuel Ernesto Garcia

Hey Hunters 👋

Today I'm really happy to share with you SidekickBar, a tool I built because I was tired of losing focus by jumping to the browser for quick AI queries. Most apps don’t have assistants built in, so I wanted something system-wide, fast and always there.


As someone who works on both Linux and Windows, I needed something lightweight and cross-platform. I started experimenting with integrating Copilot into a standalone UI, but quickly realized I needed more flexibility. Copilot is fine for general questions, but I prefer Claude for development or Gemini when writing. So… I ended up creating a multi-agent app that lets me switch between assistants depending on the moment.


For curious developers: SidekickBar is built mostly with Electron, styled with Tailwind, and the entire UI was built with Svelte.

I’d love to hear your thoughts, feature ideas, or even edge cases you’d like to see handled.

Best regards,
Manuel

Rachit Magon

@merodriguez9112 This is a very interesting proposition! I haven't tried it right now but I'm trying to understand how do you limit usage for users using the sidekickbar, can't they overuse the models and end up actually costing you more than you charge? In pricing I do not see any X amount of queries in the free plan, Y queries in the Lifetime access etc? Or am I getting it wrong?

Manuel Ernesto Garcia

@rachitmagon. Hey Rachit! Great question—and you’re right to be curious about that. SidekickBar is designed so that each user authenticates directly with the AI service they want to use (Claude, Perplexity, Poe, etc.), using their own account. That means there’s no middle layer or API proxy run by me—I don’t route the traffic or pay for your usage.

This was intentional from the start. It avoids needing to manage usage limits or tokens on my end, and it also gives users full transparency and control. If a service has free or paid tiers, you’re interacting with that service directly just like you would in your browser. SidekickBar simply gives you a more seamless way to access it from your desktop.

Appreciate you taking the time to dig into the details! Let me know if you end up trying it out. I’d love to hear your feedback.

Rachit Magon

@merodriguez9112 That's very cool & straightforward. I wonder why don't all the other apps that use public LLMs behind the scenes don't do that. Will definitely try it out soon

Nika

Yesterday, I told one guy it would be cool to have a sidebar for his project (analysing competitors according to business idea).

Now, you have applied something similar to AI. It is like reading my thoughts! :D

Manuel Ernesto Garcia

@busmark_w_nika That’s awesome, Nika! I love that timing. it’s like we’re syncing wavelengths across projects 😄.

Your idea sounds great too a sidebar for competitive analysis could be incredibly handy. Thanks for the kind words :)

Shahriar Hasan

How does SidekickBar manage context switching between multiple AI agents while keeping the user experience smooth and consistent?

Manuel Ernesto Garcia

@shahriardgm Hey Shahriar! Great question. Right now, each assistant in SidekickBar runs in its own isolated context, so they don’t share state or data between them (yet). In the future, I’d love to explore adding a shared layer that lets them “talk” to each other or share some data.

For now, the app interface adapts based on the assistant you’ve got active. Switching between them is really quick, either via keyboard shortcut or directly from the UI, so you can keep moving without breaking your flow.

Thanks for checking it out!

Van de Vouchy

How are you addressing concerns about SidekickBar’s higher RAM and CPU usage as an Electron app, especially for users on lower-end machines?

Manuel Ernesto Garcia

@vouchy That’s a really fair concern, Van and as you know electron is not the best one on this hahaha.

Right now, it’s a bit of a balancing act. SidekickBar prioritizes responsiveness, so assistants are spun up in the background to keep interactions snappy. Naturally, that means memory usage can increase depending on how many assistants a user has running in parallel. There’s no strict limit imposed, the resource use really scales with how you choose to use it.


This first version has fairly basic optimization. For future releases, maybe we can consider adding smarter memory management, like suspending inactive assistants after a period of inactivity or based on available system RAM. That would free up resources but comes with the tradeoff of slower wake times when reactivating those assistants.


Thanks again for raising it, feedback like this helps a lot!

Sam @CRANQ

I'm loving SidekickBar for keeping me in flow instead of constantly tab-switching lol! Plus w/ the different AI assistants available system-wide w/ simply a hotkey is exactly what my workflow needed!! Great work creating this cross-platform solution, Manuel!

Manuel Ernesto Garcia

@cranqnow Thank you so much Sam! That seriously means a lot. Keeping the flow intact was exactly the motivation behind building it. I’m glad the hotkey are fitting into your workflow the way I hoped they would.

If you ever have ideas for features or run into anything you think could be smoother, I'm all ears. Appreciate you taking the time to share this!

Erliza. P

🧠🦾 SidekickBar is live! Multi-agent AI right at your side — it’s like having a genius team in your toolbar 💡🔥

Nandini Gupta

I absolutely love the idea! I'm sure it'll make me feel like a superhero. I’m curious about how you handle updates as these AI models evolve so quickly? Do users automatically get access to the latest versions of tools like ChatGPT, Claude, Gemini, etc.? Excited to see how this improves daily productivity!

Manuel Ernesto Garcia

@nandini_gupta7 Thanks, Nandini! I'm really glad you liked the idea—it should feel pretty empowering! 😊

As for the updates, yes, most AI models update automatically when new versions are released. That said, if there’s a major change under the hood, there might be the occasional compatibility issue that I'd need to adjust manually. But those cases are rare and usually easy to fix.

Super excited to see how this all plays out and helps boost our daily workflows too!

12
Next
Last