OpenParable
Skip to content
← Back to skills

Claude Bowser

Apps & Automation
claude-bowser

Claude Bowser is an Agent Skill for apps & automation published in the disler/bowser repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Observable browser automation using Chrome MCP tools. Use when you need to browse websites, take screenshots, interact with web pages, or perform browser tasks in your current Chrome. Keywords - browse, screenshot, browser, chrome, bowser, ui testing, observable.

View on GitHub ↗
QUICK START

How to use this skill

Bring this guide into your coding agent with a prompt tailored to the tool you use.

  1. Open your project in Codex.
  2. Copy and paste the prompt below.
  3. Review files and risks before approving any action.
Prompt to paste
Inspect the Agent Skill at https://github.com/disler/bowser/blob/HEAD/.claude/skills/claude-bowser/SKILL.md and all supporting files. Report provenance, permissions, scripts, dependencies, and security risks. Propose an exact installation into .agents/skills/claude-bowser, but wait for my approval before writing files or running scripts.

Copying this prompt does not install or run the skill. Review third-party files before use.

Claude Bowser

Purpose

Automate browsing using Chrome MCP tools (mcp__claude_in_chrome__*) available when Claude Code is started with --chrome. This uses your real Chrome browser — observable, with your existing profile, cookies, and extensions.

Pre-flight Check

Before doing anything, verify Chrome MCP tools are available. Look for tools matching mcp__claude_in_chrome__*.

  • If available: proceed with the workflow.
  • If NOT available: stop and reply to the user: "Chrome tools are not available. Please restart Claude Code with the --chrome flag: claude --chrome"

Workflow

  1. Resize the browser window to 1440x900
  2. Execute the user's request using Chrome MCP tools (navigate, click, screenshot, etc.)
  3. Report results back to the user

Limitations

  • No parallel instances. All Chrome MCP connections share a single Chrome extension controller. Only one bowser task at a time.
  • Observable only. This uses your real Chrome — there is no headless mode.