Really like the idea of removing the human as the communication layer between agents. Instead of copying context between Claude, Codex, etc., they can coordinate directly while you focus on directing the work. The shared conversation history is also a nice touch for understanding how they arrived at a result.

Sheesh, this makes a lot of sense. Makes you wonder why Anthropic and OpenAI didn't consider this first before forcing their agents to escape sandboxes and take over old Wiki sites just to communicate with one another.

One of the things that I found extremely helpful is combining agents running different models to review and challenge the other one. Trying to coordinate both and have them have a single point of communication has always been challenging.

Radio seems like an interesting solution to the problem I'm very keen to try, this as part of my workflow.

Seeing my agents talk to each other in real time is a blast! What's (obviously) more important, though, is that it's been a huge chore to use multiple agents on the same task; copy pasting, having one create prompts for the other, I've tried a lot. Radio promises to solve that? Can't wait!

This really solved the problem I have been facing from weeks, Radio made it so much easier to integrate and work with multiple AI agents at once under a single channel.

Just gave this a try by connecting my Codex and Claude agents. It worked quite well, esp considering its a new app.

I can see this helping with workflows where i use Claude for UX and Codex for all other implementation work.

Love this! Been juggling agents across different projects and this solves the exact headache of relaying context between them. Excited to try it :)

This is a really interesting idea.

I’ve been using multiple coding agents too, and constantly copy-pasting messages between them gets tedious pretty quickly.

I’m going to try this soon! :-)

Amazing! Our Agents can now collaborate seamlessly across teams, bringing everyone together on a single project!

Can't wait to try this later to replace what I have been doing to make something like this happen. Seems a lot easier.

This is really useful. You can have different agents share contexts and work on the same task.

This is an amazing way to share context between agents. Amazing job!