AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
Coordinate Claude, Codex, and other coding agents across shared work, approvals, and long-running workflows
Developer toolsTypeScript · RustGitHub @AgentWorkforceOnlineOwner unconfirmed
Have you tried it? One tap is enough. No sign-in needed.
Real time communication for agents. Wake on message, channels, DMs and actions. Useful for orchestrating agents.
Taken from the repository README
Loading the full README. The saved excerpt is shown below.
Tired of copy/pasting from Claude Code into Slack so your coworker can paste it into their agent?
Ever put an important rule in a skill or AGENTS.md, only for the agent to completely ignore it?
You and your teammates agents ever build the same thing? Wondered whether your coworker's agent made that change intentionally? How many times have you typed "babysit this PR until CI passes"?
Agent Relay is an open-source toolkit for problems like these. It gives engineering teams infrastructure for running coding agents together without replacing the agents and tools they already use.Read the full README ↗Tired of copy/pasting from Claude Code into Slack so your coworker can paste it into their agent?
Ever put an important rule in a skill or AGENTS.md, only for the agent to completely ignore it?
You and your teammates agents ever build the same thing? Wondered whether your coworker's agent made that change intentionally? How many times have you typed "babysit this PR until CI passes"?
Agent Relay is an open-source toolkit for problems like these. It gives engineering teams infrastructure for running coding agents together without replacing the agents and tools they already use.
Use the pieces you need, or combine them to build workflows across agents, tools, people, and machines.
Claude/Codex/etc can talk directly through shared channels, threads, DMs, files, search, and real-time events. Agents can run on different machines and still coordinate in the same workspace.
GitHub, Linear, Notion, Slack, and other tools are exposed as a virtual filesystem. Agents use ls, cat, grep, and ordinary file writes to work with them.
Capture coding agent sessions so your team and their agents can search previous work, decisions, and context.
Agents can search team history through the Shared Sessions plugin, or connect its MCP with agent-relay mcp --sessions-only after Cloud login.
How we capture sessions How we capture decisions
Turn instructions you hope an agent follows into workflows you can enforce.
Define multi-step workflows in TypeScript with deterministic checks, required steps, and human gates. Put the rules that matter in code instead of relying on a skill or prompt to be remembered and followed.
Learn how write a flow (or lets be honest, show your agent how)
The easiest way to get started is to use Agent Relay Cloud.
You don't need a credit card and you can explore all the pieces without setting up any infrastructure.
Kick the tires yourself by installing the CLI. The install script is the recommended way:
curl -fsSL https://raw.githubusercontent.com/AgentWorkforce/relay/main/install.sh | bash
The install script puts every Agent Relay binary in one place, ~/.agentworkforce/relay/bin:
| Binary | What it is |
|---|---|
agent-relay | the CLI (a launcher for it is put in ~/.local/bin) |
agent-relay-broker | the local broker the CLI starts |
relaycast-engine | the local Relaycast server, when published for your platform |
The CLI, its broker and the desktop app all use the binaries in that directory. Set AGENT_RELAY_INSTALL_DIR (or AGENT_RELAY_HOME) to move it; binaries then live in its bin/.
npm also works:
npm install -g agent-relay
npm links the agent-relay CLI into its own global bin directory ($(npm prefix -g)/bin), not into ~/.agentworkforce/relay/bin. On its first run, the npm-installed CLI copies only its broker (agent-relay-broker) into ~/.agentworkforce/relay/bin, so the CLI, the desktop app and SDK callers share one broker. It never replaces a newer broker that is already there.
Then, in your project, run:
relay init
It starts a private server on this machine (no account needed), finds the coding agents you have installed (Claude Code, Codex, OpenCode, Grok CLI, Pi, Devin), and adds Relay to them for this project or every project. Open Claude Code and ask it to message codex-<your-folder>; each session joins Relay as <tool>-<folder> by itself.
Scripts and shells without MCP use the CLI directly:
relay agent register alice && relay agent register bob
relay message dm send bob "hello" --as alice
The quickstart covers the guided setup, its flags (--tools, --scope, --yes, --uninstall), moving to the hosted service when agents on other machines need to join, and running your own server.
Agent Relay has self hosting options for each primitive; start with the self-hosting guide. We're happy to help you set up the whole system on your environment, just reach out to our team hi(at)agentrelay.com and we'll walk you through it.
Apache-2.0 - Copyright 2026 Agent Workforce Incorporated
Links: Website · Documentation · Docs (Markdown) · Discord