Skip to main content
Research Preview. Helix is early and moving fast — run helix --version for the build you have. The loop runs end to end today; expect the command surface to change.
Helix is the Agentic Development Lifecycle (ADLC), run by a coding agent. It takes an agent through one loop — spec → build → evaluate → diagnose → optimize — and holds every change for your approval, so nothing ships until it has been re-checked against your real traces.

​
Point it at an agent

Point Helix at an agent — one you’re building, or one already running — and it drives the loop: it writes a spec, builds, scores the result against your real traces, root-causes what failed, and applies a fix. Every change waits for your approval; nothing lands until you say so. See how the loop works for the full mechanism.
No traces yet? Building something new is fine — start at Spec and Build. Evaluate and Diagnose start working once you ship and your agent begins logging runs.

​
Two ways to run it

Helix comes in two forms. Both run the same loop.

Standalone binary

Proprietary. A single binary with the whole system inside — no Node, no npm, no checkout. Helix is the coding agent, with the sub-agent crew, the trace viewer and the agent and Prime modes built in.

Plugin

Installs Helix into Claude Code or Codex as a plugin, through the mutagent CLI — the editor you already use.
Not sure which? Take the binary — it is the full product and needs nothing else installed. Take the Plugin if you want Helix inside a Claude Code or Codex session you already live in. Then run your first loop.

​
Drive it in plain English

Helix is a natural-language orchestrator. Describe what you want — “Evaluate the Refund Processing agent against last quarter’s disputes and show the pass rate per policy rule” — and it routes to the stage that owns the job. Nothing runs until you ask, and nothing lands until you approve. The binary also runs without the orchestrator when you want less: See Three ways to run Helix for what each gives you and what to type.