Skip to main content

Install & setup

Start hereOne-line installer, prerequisites, and credentials in three steps.

Quickstart

5 minutesRun init → simulate → optimize on your own agent project.

All commands

ReferenceEvery relai command with flags, examples, and failure modes.

The loop, at a glance

Everything in RELAI maps onto a four-step cycle that repeats. You can run it by hand, or drive it from your coding agent with a plugin. For the full picture — an illustration and each component’s role — see The learning loop.
1

Initialize

From a git-tracked agent project, relai init seeds and locally validates a simulator harness before registering the agent. Paused runs continue with relai init --resume.
2

Author environments & evaluators

Capture behavior to learn or preserve as repeatable learning environments — from a prompt, or from a run log plus feedback. Optionally register benchmark suites.
3

Simulate

relai simulate measures how the agent behaves today against the environments or benchmarks you select.
4

Optimize

relai optimize proposes or applies improvements — and can open a PR with the changes.
New here? No agent of your own yet? Try the sample agent — walk the whole loop in your browser, no install required.