amoeba Download

Multiplayer coding with AI agents

Guides, explainers and release notes from the Amoeba team: shared sessions, collision prevention, local-first architecture, and running any agent harness in one IDE.

Guide6 min

How to stop AI coding agents from editing the same file

Why parallel agents collide on shared code, and how line-level collision prevention names the teammate and function before tokens burn.

KKenneth
Explainer5 min

Multiplayer coding with AI agents: one shared session for the team

What a shared agent session is, how live cursors and approvals work, and why it beats everyone running Claude or Codex alone.

KKeyu
Guide7 min

Claude, Codex, OpenCode or ACP: running any agent harness in one IDE

A guide to bringing your own subscription or API key, connecting ACP harnesses like OmniCode and Hermes, and handing tasks between them.

KKenneth
Architecture8 min

Local-first AI coding: how Amoeba shares sessions over WebSocket

Log in from the CLI, run a local server, and let teammates join over WebSocket. Your code never leaves your machines.

KKeyu
Security5 min

Approval gates: keeping AI agent changes off your branches until a human signs off

Roles, per-repository permissions and an approval step before any agent push. How enterprise teams run agents without giving up control.

KKenneth
Explainer6 min

Shared project memory: giving every AI agent the same context

Decisions, conventions and context pinned to files in your repo, readable by every agent, and flagged when the code moves.

KKeyu
Guide4 min

Parallel subagents that report back as one result

How one agent splits a task into subagents, runs them side by side, and returns a single diff for review.

KKenneth
Tutorial3 min

Amoeba CLI quick start: connect a repo and open a shared session in five minutes

Install, sign in, connect a repository, open a session and invite a teammate. The whole first run, command by command.

KKeyu
Security5 min

Sandboxed AI agents with zero data retention

Each agent works in an isolated worktree it is allowed to break. Nothing is retained, and nothing lands without review.

KKenneth
Explainer4 min

Hit your AI usage limit mid-task? Hand the agent to a teammate

When a provider limit is reached, Amoeba offers to pass the live task, with its prompts and diff, to someone with capacity.

KKeyu