Goalrail

Goalrail

An open layer over Claude Code, Codex, Pi, and the agents you write yourself: switch harnesses without rewriting, keep work inside policies and sandbox boundaries, and collaborate on the same live session from any device.

$ curl -fsSL https://goalrail.dev/install.sh | sh
Get Started macOS App @goalrail
Goalrail web UI running two sub-agents in parallel on a shared session

Core workflows

  • Multi-agent workflows: coordinate coding, review, and debate loops without locking the workflow to one model provider.
  • Contextual Policies: stateful spend caps, model routing, and risk-based escalation.
  • Secure OS Sandbox: restrict filesystem and network access. Hide credentials from the agent, and broker access to them. Run YOLO mode safely.

What Goalrail gives you

Composition

Combine multiple models, harnesses, and techniques without rewriting code. Switch between Claude Code, Codex, Pi, and your own agents with one-line changes.

Architecture

A runner wraps any agent in a sandboxed, uniform session. A server adds policies and shared history, and exposes every session over the terminal, the web, a native app, mobile, and a REST API.

Goalrail architecture: CLI agents and custom agents run through a runner (on your machine, Modal, or Daytona), then a server that adds policies and history, reachable from a terminal, the web, native and mobile apps, and a REST API.

Goalrail is alpha and built in the open. Try it and give us feedback in Telegram.

@goalrail