Today, we're launching the Traycer Desktop App. Free. Open source. Built with AI orchestration in mind. What’s new: • Bring your existing subscription (Claude, Codex, Opencode, etc.) • Agent-to-Agent communication and Loops • Persistent workspaces with tabs and sub-tabs • Share tasks and collaborate with teammates Capability stopped being... the hard part. The environment is the new frontier, and that’s where Traycer is.show more

Install now at :

i tried it today, UI is amazing.....

this is insanely cool

amaziiinnnnnnnggggg

@GillTanveer89 Congratulations, looks killer!

hey @TraycerAI are you guys going to be building a mobile app?

@GillTanveer89 Congratulations guys

I mainly want the button that says: show me exactly what this thing is about to touch.

Really like the UI

Amazing! I will try!

把现有订阅接进来这点还挺省事

Love this release!! I did however immediately fork it and add remote hosts over tailnet (pls add official remote hosts soon so I dont have to maintain a fork 🙏)

Congrats 🎉

Very interesting!

Just downloaded. This is absolutely awesome! Kudos gang!

Persistent workspaces matter more than people admit. My tabs are not emotionally stable.

@barre_of_lube bro

lmao Im gonna just stop building it cause I literally been doing almost the same but nowhere near at this level yet. Can it use Copilot?

Yes, we support Copilot!

Trying it today at the office nice

Congrats!

Interesting, interesting, is it possible to set an overall goal and see wich Agent/Model is taking which role?

There are several skills packaged out of the box for collaborative planning, brainstorming, and agent-to-agent debate. /execute implements the changes end-to-end using multiple agents, reviewers under the oversight of an orchestrator agent. You can define the overall agent selection criteria at the user level, repository level, or within the prompt: e.g., use Claude Code, Opus with high reasoning for implementation, and Codex, GPT-5.5 for review

Where is this opensource? Link? @traycer_ai

i'm unclear of the value prop after reading your website but it's undeniable your reviews are strong and your ui is excellent so i'll try it out and figure out how it helps my app ecosystem along the way

congrats on the launch. persistent workspaces plus agent-to-agent loops feels like the right place to make orchestration less invisible.

Uhh i might be kinda dum dum but how do i switch workspaces? seems like the interface that should let me do it doesn't let me do it (can't click the workspaces)?

"Free, open-source, and built for agent-to-agent loops—this is exactly how AI orchestration should be done. Huge win for the developer community! 💻🚀"

The agent-to-agent loop is the interesting piece. Do you keep a shared receipt of who changed what between agents, or does each workspace own its own trail?

help, claude only (opencode works) returns this on every prompt: Claude's API returned an unexpected error before the turn could finish.Errorunknown

so cool, building the exact same thing at @anymo_ai

Traycer Desktop looks incredible — open source + BYO subs + persistent workspaces is a huge win!

bring your own sub is the only move that matters. desktop wrapper is just chrome around it

This is huge Free, open source, agent loops. Traycer just changed the game

Every day I wake up and there's a new AI orchestration app

É bom??

Firstly Congratualtions ! But as much as like to work with AI orchestration , i am starting to feel its really slowing down the time to output when i am compare with codex or code . I also miss the genreal harness from the native LLm env