Agent Browser

Agent Browser is an Automation skill for Claude Code, published by fcakyon in claude-codex-settings.

1.1K stars102 forkson fcakyon/claude-codex-settingsAdded 2026/08/20Repository updated 2026/08/20

ai-agentsai-toolsclaude-aiclaude-codeclaude-code-pluginclaude-skillsclaudecodeclaudecode-configcodexcodex-clicommit-messagecursordokployfrontend-designgemini-cligithubhetznerkimi-k3pluginpull-requests

Install in seconds

Install Agent Browser

Copy Agent Browser into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.

terminal

npx degit https://github.com/fcakyon/claude-codex-settings/tree/main/plugins/agent-browser/skills/agent-browser ~/.claude/skills/agent-browser

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/fcakyon/claude-codex-settings.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file: plugins/agent-browser/skills/agent-browser/SKILL.md in fcakyon/claude-codex-settings
Installs to: ~/.claude/skills/agent-browser
Collection: One of 25 skills cataloged from this repository
Category: Automation — 1523 skills

What Agent Browser does

Agent Browser is a fast browser automation CLI that uses Chrome/Chromium via CDP and accessibility-tree snapshots so agents can navigate, click, fill, extract, screenshot, and manage tabs with compact element refs. Use it when a user asks to interact with a website, fill a form, scrape data, log in, or automate browser tasks.

Agent Browser is cataloged under Automation on DirSkills. Agent Browser comes from a repository tagged ai-agents, ai-tools, claude-ai, claude-code and claude-code-plugin.

Documentation

README

Excerpt

agent-browser core

Fast browser automation CLI for AI agents. Chrome/Chromium via CDP, no Playwright or Puppeteer dependency. Accessibility-tree snapshots with compact @eN refs let agents interact with pages in ~200-400 tokens instead of parsing raw HTML.

Most normal web tasks (navigate, read, click, fill, extract, screenshot) are covered here. Load a specialized skill when the task falls outside browser web pages — see When to load another skill.

The core loop

agent-browser open <url>        # 1. Open a page
agent-browser snapshot -i       # 2. See what's on it (interactive elements only)
agent-browser click @e3         # 3. Act on refs from the snapshot
agent-browser snapshot -i       # 4. Re-snapshot after any page change

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Agent Browser

What else does fcakyon publish alongside Agent Browser?
Agent Browser is one of 25 skills that DirSkills catalogs from fcakyon/claude-codex-settings, the repository it ships in. Its siblings there include ADHD Output Style, Azure Tools Setup and Azure Usage. Each one is a separate skill with its own page in this directory, installs the same way Agent Browser does, and is maintained by fcakyon in that same repository. The rest of the collection is listed on the fcakyon/claude-codex-settings page.
How does Agent Browser compare to other Automation skills?
Agent Browser ranks #774 by stars among the 1523 Automation skills in this catalog. The most-starred ones next to it are Autonomous Loops, Autonomous Agent Harness and Automation Audit Ops. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Agent Browser against them. Open each page to compare what they document and how they install.