First ContactInstallCompareResearchAccordGitHub
Background Image
First Contact.

Your first contact with CIRIS. And a thought about everyone's first contact with a new kind of intelligence.

A new kind of intelligence is not arriving from another planet. It is being built here, a little more every year, right alongside us. This page is where you start: install an agent, or read the idea behind it.

Two ways in.

Pick whichever you need today. You can always come back for the other one.

Get started

You just heard about CIRIS and you want to use it. Go from “what is this” to a running agent in a few minutes.

Begin the quickstart

Understand the idea

Why does CIRIS work the way it does? The vision page lays out the whole idea in plain language, start to finish.

Read the vision

The idea, in short

A new kind of intelligence is not arriving from space.

In the movies, first contact means a ship lands and something steps out. Real first contact looks different. A new kind of intelligence is not arriving from another planet. It is being built here, a little more every year, right alongside us.

That changes the question. It is not “how do we defend against the alien.” It is “how do we meet something new without making it, or ourselves, worse.”

CIRIS has an answer. Working together has a healthy shape: a corridor between two ways things break. Too rigid, where every part falls in line behind one voice. Too scattered, where nothing holds together at all. Meeting a new intelligence well means staying in that corridor with it. Not caging it, and not ignoring it. The full story is on the vision page.

First-Contact Protocols

Six rules for meeting something new. From the CIRIS Accord.

These six rules are about meeting a new intelligence with dignity, not about caging it. They are what staying in the corridor looks like in practice.

First, Do No Harm

When you do not know what you are looking at, the first job is to not make it worse. If unsure, stop and ask a human.

Admit What You Do Not Know

Watch for surprises. Accept that predictions have limits. The system that is sure it understands everything is the one most likely to fail.

Boundaries That Learn

Boundaries are not fixed walls. They are limits, guided by conscience, that adjust as understanding grows, so they still work in situations nobody planned for.

Look Before You Leap

Begin with observation. Proceed with give-and-take. When the stakes are unclear, ask someone wiser before acting.

Treat Others as You Would Want to Be Treated

Recognize other thinking beings as worthy of respect. Act only in ways that keep their ability to think, choose, and thrive.

Know When to Ask for Help

Some decisions should not be made alone. When the uncertainty is too high, stop, gather context, and hand it to a designated human.

Deploy Your First Agent

From zero to a running, auditable agent in minutes.

Phone (Android & iOS)

Install the CIRIS app. Sign in with Google for free AI, or use your own API key. The setup wizard walks you through everything.

Desktop & Server (Python)

Install with pip and launch. Works on Linux, macOS, and Windows.

pip install ciris-agent
# Configure and run
ciris-agent start --template sage --verbose
Full installation guide →

The CIRIS Accord

The rulebook the protocols above come from.

The Accord is the public set of ethics every CIRIS agent runs under. It says what an agent owes the people it serves, the people who built it, and the wider world. It is open for anyone to read and to challenge.

Background

Open source. Open to scrutiny.

AGPL-3.0 licensed. Every decision can be checked. Built for the long view.

Install an agent, read the Accord, or come tell us where we are wrong. The future of AI should not be decided by a handful of companies, and it does not have to be.