Core Agent Browser

This is an internal automation layer that wraps the agent-browser CLI for browser workflows. It's used under the hood by rust-learner and other research skills, not something you'd typically invoke directly. The docs are clear about this: reach for rust-learner or actionbook first, and only drop down to this for interactive testing or when you need screenshots and form filling on sites without pre-computed selectors. The command reference is solid, snapshot with the -i flag gives you element refs like @e1 that you can click or fill, and there's a sensible wait system. Useful to know it exists if you're debugging browser automation, but treat it as a last resort.

skills.sh installs: 2.2K
GitHub stars: 1.6K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add actionbook/rust-skills --skill core-agent-browser --agent claude-code

Installs into .claude/skills of the current project.

Files

Browser Automation with agent-browser

Priority Note

For fetching Rust/crate information, use this priority order:

rust-learner skill - Orchestrates actionbook + browser-fetcher
actionbook MCP - Pre-computed selectors for known sites
agent-browser CLI - Direct browser automation (last resort)

Use agent-browser directly only when:

actionbook has no pre-computed selectors for the target site
You need interactive browser testing/automation
You need screenshots or form filling

Quick start

agent-browser open <url>        # Navigate to page
agent-browser snapshot -i       # Get interactive elements with refs
agent-browser click @e1         # Click element by ref
agent-browser fill @e2 "text"   # Fill input by ref
agent-browser close             # Close browser

Core workflow

Navigate: agent-browser open <url>
Snapshot: agent-browser snapshot -i (returns elements with refs like @e1, @e2)
Interact using refs from the snapshot
Re-snapshot after navigation or significant DOM changes