Opensourcing Multiplayer AI in Discord

mehdim1 pts0 comments

bunny: build in multiplayer with humans and agents on a shared environment

GitHub

Discord

Collaborative development for the AI era

From chat to shipping, keep humans and AI agents in the same context.

Turn a VM or Docker container into a shared dev station with shared shells, live previews, and chat-native workflows. Self-hosted by default.

Discuss in chat. Execute in shared context.

View on GitHub

Read the docs

Explore the vision

Shipping in a single thread

code review, bunny AI agent (separate paths) -->

bunny -> GitHub (keep separate from review line at y=402) -->

bunny -> browser -->

bunny, then bunny -> thread -->

Thread

Can we ship today?

Login is flaky in prod

@bunny launch live stream preview

We should change the color<br>and the layout

Yes, I agree, and we<br>should propose social login

@bunny stop live preview

@bunny fix and open PR

Review looks good

Validate

Click me

Reject

PR merged to main

AI agent

GitHub

gated push

Code review<br>in Thread

- login(email, password)<br>+ loginWithOAuth(provider)<br>return session.create(user)

Can we ship this change?

OAuth flow looks correct

🤖 Code review validated<br>by AI agent

Code review done

live preview

Sign in

🧠

Replay

Same moment, one source of truth

Without bunny

Dev A<br>ssh

Dev B

team chat

deploy

Terminal

Cursor

GitHub

main?

v2?

two actions · one conflict

vs

With bunny

Dev A<br>ssh

Dev B

team chat

deploy

ssh · VM<br>bunny

Term.<br>Cursor<br>Claude<br>Codex<br>GitHub<br>Supabase

✓ synced

two actions · one state

Governed gateway, shared context

Without bunny

Dev A

Dev B

IDE

team<br>chat

Claude

Cursor

Codex

GitHub

Supabase

other tools<br>& services

fix login

deploy?

ungoverned · context lost<br>each tool, its own memory

vs

With bunny

Dev A<br>ssh

Dev B

IDE

team<br>chat

bunny<br>ssh VM

Claude

Cursor

Codex

GitHub

Supabase

self-hosted tools

managed services

fix login

deploy?

@bunny do it

governed gateway · shared context in memory<br>chat ↔ bunny ↔ every tool

Earlier proposals<br>Current version

No shared context

team chat

Terminal

Browser

Codex

Git<br>separate contexts

vs

Shared context

team chat

bunny

Terminal

Browser

Codex

Git<br>one shared context

Proposal: 2 devs and silos

Broken context

v2?

Dev A

main?

Dev B

fix login<br>team chat

login?

Terminal

broken UI

Browser

v2 ≠

Codex

to merge

Git

fix login

fix login

fix login

fix login

fix login

fix login

same topic · conflicting versions

vs

Relevant context

shared context

discusses<br>discusses

commands

@bunny run tests

@bunny open preview

@bunny commit fix

✓ fix login

✓ fix login

control plane

✓ fix login

✓ fix login

✓ fix login<br>source of truth

✓ fix login

✓ fix login

✓ fix login

✓ fix login

Dev A

team chat

tests failing?

login broken?

on it

@bunny run tests

Dev B

bunny

Terminal

Browser

Codex

Git<br>discuss in chat · control tools · shared context

The shift<br>From local repos to shared, executable context

Instead of each developer working locally and synchronizing through GitHub, teams collaborate<br>directly around a common remote environment, with chat, roles, and permissions intact.

Shared remote environment

A VPS or container becomes the team's persistent workspace: terminals, live previews,<br>streamed browsers, and a unified timeline where code, feedback, and experiments converge.

Chat-native workflows

Work from a channel with the same context, the same coding agent, and distinct<br>rights for every contributor, whether engineers, designers, operators, or non-technical members.

Open & self-hosted

Install the tools you already use (shell, CLIs, scripts, Codex, Claude) through prompts<br>and shared workflows in an environment your team controls.

Beyond code<br>Versioning that captures how software is actually made

The Git commit is often the only durable artifact explaining a project's evolution. But in<br>modern workflows, meaning lives elsewhere: chat threads, agent prompts, collective decisions,<br>tests run, errors encountered, and trade-offs accepted.

bunny adds a semantic layer on top of code versioning, connecting each important change to<br>the discussions, goals, and interactions that led to it.

Chat discussions

AI agent prompts

Decisions & trade-offs

Tests & regressions

Code changes

One collaboration graph, not just a commit log.

GraphRAG<br>Queryable context for every change

Enriched metadata is ingested into a GraphRAG, enabling questions that Git alone cannot answer.

Why was this commit made?

Which discussion led to this implementation?

What objective did this change serve?

Which agents or contributors influenced this decision?

In the long term, the remote environment becomes the guardian of the repository. Git doesn't<br>disappear. It stops being the project's only mental model.

Parallel validation<br>CI feedback at the speed of vibe coding

Current workflows are poorly suited to AI-assisted development. Teams iterate faster, rely<br>more on agents, yet CI arrives too late, takes too long, and sometimes gets bypassed...

bunny login chat context shared team

Related Articles