Claude Code

Use Claude Code with Steel CLI to drive real browser sessions, scrape rendered pages, and monitor browser workflows from the terminal.

Overview

The Claude Code integration uses Steel CLI as a terminal tool through agent skill steel-browser. This lets Claude Code:

  • Start and control Steel browser sessions from the Claude session
  • Scrape fully rendered pages and perform computer-use actions
  • Inspect live browser state with snapshots and the Steel session viewer
  • Turn successful browser runs into repeatable scripts or project-specific workflows

Claude Code already works well with local tools and shell commands. Steel extends that model with a real cloud browser for sites that need JavaScript, session state, or interactive navigation.

Requirements

  • Node.js: Version 18 or higher
  • Claude Code: Installed locally
  • Steel CLI: Installed locally
  • Steel API Key: Active Steel account

Setup

Step 1: Install Steel CLI

Terminal
curl -LsSf https://setup.steel.dev | sh

Step 2: Log in

Terminal
steel login

The steel-browser skill gives Claude Code better command discovery and more reliable browser workflows.

Terminal
npx skills add https://github.com/steel-dev/cli --skill steel-browser

Restart Claude Code after installing so it can discover the skill.

Once Steel CLI is available, Claude Code can use it directly:

Authenticated workflows

For authenticated sites, it is better to prepare reusable auth state in Steel ahead of time rather than asking Claude Code to handle login from scratch during every run.

See:

Live debugging

Steel sessions return a viewer URL that makes it easier to monitor what the agent is doing in real time. This is useful when Claude Code reaches a modal, a sign-in wall, or a page that does not behave as expected.

Repeatable workflows

After a successful run, Claude Code can help turn the browser workflow into something more repeatable:

  • A bash script for local execution
  • A project-specific command or workflow
  • A documented runbook for recurring tasks

Constraints

  • Command approvals depend on your Claude Code settings. Shell access may require approval depending on your permission mode.
  • First runs are usually the roughest. Dynamic web apps often need a few retries before the workflow is stable.
  • Authenticated sites work best with prepared Steel auth state. Reusing profiles or auth context is generally more reliable than repeated interactive logins.

Additional Resources