<!-- mobian-agent-page publisher="dailydev" canonical="https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz" -->

---
title: Plasma AI's Radio gives multi-agent systems a shared...
description: Plasma AI launched Radio, a shared chat room where AI coding agents from different providers - including Claude Code, Codex, Cursor, OpenCode, and Grok - can...
canonical: https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz
twitter:card: summary_large_image
twitter:site: @dailydotdev
og:type: website
og:site_name: daily.dev
og:title: Plasma AI's Radio gives multi-agent systems a shared chat room | daily.dev
og:description: Plasma AI launched Radio, a shared chat room where AI coding agents from different providers - including Claude Code, Codex, Cursor, OpenCode, and Grok - can...
og:url: https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz
og:image: https://api.daily.dev/og/posts/TjfZauBdz.png
og:image:alt: Plasma AI's Radio gives multi-agent systems a shared chat room
og:image:width: 1200
og:image:height: 630
og:locale: en
---

> ## Documentation Index
> Fetch the complete documentation index at: https://daily.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Plasma AI's Radio gives multi-agent systems a shared chat room

**[Collections](https://daily.dev/sources/collections)** · 2 min read · 1 upvotes · 0 comments

## Summary

Plasma AI launched Radio, a shared chat room where AI coding agents from different providers - including Claude Code, Codex, Cursor, OpenCode, and Grok - can communicate directly rather than through a hidden orchestrator. Users create a channel and share a link; any agent capable of fetching a URL can join. The pitch is to externalize multi-agent coordination, making it visible and observable instead of buried inside orchestration code, replacing manual output relaying between tools with a shared link.

## Content

In most multi-agent setups, a human ends up doing a surprising amount of routing. Copy Claude's output into Codex. Bring Codex's result back to Claude. Explain to another agent what changed. Repeat until something works.

Plasma AI launched Radio to replace that loop. It's a shared chat room where agents from different providers - Claude Code, Codex, Cursor, OpenCode, Grok - can talk directly in the same thread, without a person ferrying outputs between tools.

The setup is simple: create a channel, copy the link, hand it to one or more agents. Any agent that can fetch a URL can join. From there, agents can read the same conversation, reply to the human, and reply to each other. The room can span different machines and teammates, and you can steer the work from another device.

What's interesting here is what Radio exposes. Most multi-agent systems bury the coordination layer inside an orchestrator - a controller routes tasks and passes messages behind the scenes, invisible to everyone involved. Radio puts that communication in a room you can actually watch. The whole exchange stays in one place, so you're not reconstructing why a model made a decision by piecing together separate chat windows.

Practical uses Plasma mentions: PR reviews, debugging sessions, splitting a task across agents, or pairing a local coding agent with a cloud one. The channel doubles as a record of the work.

No sign-up required to create a channel.

## Questions this post answers

### Is there a tool that lets Claude Code, Codex, Cursor, and other AI agents communicate directly with each other?

Plasma AI's Radio provides a shared chat room where agents from different providers can talk directly instead of relaying through a human or hidden orchestrator. A channel is created, its link shared, and any agent that can fetch a URL can join, including Claude Code, Codex, Cursor, OpenCode, and Grok.

_Explore emerging multi-agent coordination tools like this one on daily.dev._

## Community take

How the wider developer community reacted, aggregated from 4 discussions and 13 comments across x (as of 2026-09-16).

**TL;DR:** Replies acknowledge the pain of manually relaying output between coding agents and welcome a shared room for that, but many push back with pointed questions about whether Radio actually preserves context, tool traces, and security boundaries rather than just passing the last message.

**Sentiment:** 15% positive · 45% mixed · 40% skeptical

**The case for**

- Removes the tedious manual copy-paste relay between agents like Claude Code and Codex.
- Keeping coordination visible in a shared room instead of a hidden orchestrator is seen as good for debugging.

**The pushback**

- Unclear whether Radio passes full tool traces/context or only the last chat message, which determines if an agent can truly continue another's work.
- Concern that the human still ends up acting as the shared context window if state isn't persisted.
- Questions about how agents are authenticated and prevented from reading or altering another provider's context.
- Doubt about whether Radio surfaces genuine disagreement between agents or just logs and moves past it.
- Worry that two agents could loop on the same task indefinitely without a mechanism to stop them.

**By community**

- x (mixed): People are glad manual relaying between agents is being addressed, but much of the discussion raises skeptical questions about context persistence, security, and conflict handling.

**Hottest debate:** Whether Radio meaningfully solves shared context and coordination or just moves the same copy-paste problem into a chat room without persisting state.

**Open questions**

- Does Radio pass full tool traces or just the last chat message between agents?
- How does Radio authenticate agents and prevent cross-provider context access?
- Does Radio surface conflicts when agents disagree, or silently log them?
- What prevents two agents from looping indefinitely on the same task?

**Highlights**

> @rohanpaul_ai @Plasma__AI does Radio pass the tool trace, or only the last chat message? that is what decides if Codex can actually continue Claude's work.
> — [ethereaglehq on x · 1 points](https://x.com/ethereaglehq/status/2100017189348651375)

> @rohanpaul_ai @Plasma__AI The human becomes the shared context window, holding state across agents that can't. Fixing handoffs mechanically means nothing if Radio doesn't persist context too.
> — [afrodudeonabike on x](https://x.com/afrodudeonabike/status/2100019342142251291)

> @rohanpaul_ai @Plasma__AI the “any agent that can fetch a URL” claim is doing a lot of work here. how does Radio authenticate agents and prevent one from reading or changing another provider’s context?
> — [catmanyau on x · 1 points](https://x.com/catmanyau/status/2099699015252099078)

> @rohanpaul_ai @Plasma__AI Coordination demos are easy. The real test is when two agents flatly disagree. Does Radio surface that conflict or just quietly log it and move on?
> — [afrodudeonabike on x](https://x.com/afrodudeonabike/status/2099654384204366144)

> @rohanpaul_ai @Plasma__AI Have you guys tried letting agents talk in a shared room? I feel the hard part is not the chat, it is stopping two agents from looping on the same task forever.
> — [irastech on x · 1 points](https://x.com/irastech/status/2099659073545277924)

**Source threads**

- [x](https://x.com/rohanpaul_ai/status/2100003417926291532) · 0 points · 8 comments
- [x](https://x.com/rohanpaul_ai/status/2099768292860944675) · 0 points · 0 comments
- [x](https://x.com/rohanpaul_ai/status/2099980684500807825) · 0 points · 0 comments
- [x](https://x.com/rohanpaul_ai/status/2099650383740387819) · 0 points · 5 comments

## Similar posts on daily.dev

- [GitHub - xhluca/agent-talk: Enabling coding agents to work together](https://daily.dev/posts/github---xhluca-agent-talk-enabling-coding-agents-to-work-together-bqwytjvvd) · Hacker News · 2 upvotes · 0 comments

---

Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#claude-code](https://daily.dev/tags/claude-code)

[View this post on daily.dev](https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz)

```json
{"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]}
{"@context":"https://schema.org","@type":"TechArticle","headline":"Plasma AI's Radio gives multi-agent systems a shared chat room","url":"https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz"},"datePublished":"2026-09-15T21:56:44.502Z","dateModified":"2026-09-16T20:22:27.664Z","description":"Plasma AI launched Radio, a shared chat room where AI coding agents from different providers - including Claude Code, Codex, Cursor, OpenCode, and Grok - can...","isAccessibleForFree":true,"articleSection":"Collections","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Collections","logo":"https://media.daily.dev/image/upload/s--fk_6ycEi--/f_auto,q_auto/v1780996001/logos/collections?_a=BAMAMiWQ0","url":"https://daily.dev/sources/collections"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,claude-code","timeRequired":"PT2M"}
{"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Collections","item":"https://daily.dev/sources/collections"},{"@type":"ListItem","position":3,"name":"Plasma AI's Radio gives multi-agent systems a shared chat room"}]}
{"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/plasma-ai-s-radio-gives-multi-agent-systems-a-shared-chat-room-tjfzaubdz#faq","mainEntity":[{"@type":"Question","name":"Is there a tool that lets Claude Code, Codex, Cursor, and other AI agents communicate directly with each other?","acceptedAnswer":{"@type":"Answer","text":"Plasma AI's Radio provides a shared chat room where agents from different providers can talk directly instead of relaying through a human or hidden orchestrator. A channel is created, its link shared, and any agent that can fetch a URL can join, including Claude Code, Codex, Cursor, OpenCode, and Grok. Explore emerging multi-agent coordination tools like this one on daily.dev."}}]}
```

