from qaskills97
CLI-driven browser automation with persistent multi-session support, cookie management, JavaScript execution and parallel workflows for long-running automated t
Provides agent-driven automation using the browser-use CLI: persistent named sessions, multi-session parallel runs, cookie export/import for auth persistence, JavaScript execution inside pages, tab and network control, and monitoring patterns for long-running workflows. Concrete examples and install instructions are included.
Use this skill when an agent must automate authenticated browser workflows, extract structured data via page JS, run parallel account tasks, or maintain session state across runs. Ideal for CI automation, scraping behind auth, and monitoring dashboards.
Best suited to developer/coding-capable agents (Claude Code, Cursor, Copilot-style agents) and any agent that can run CLI tools and interpret JavaScript execution results.
This skill is a lengthy reference manual for a 'browser-use' CLI tool with session management, cookie handling, and JS execution. However, the real browser-use Python package is an AI agent library — the CLI interface described here (with subcommands like 'session create', 'cookies export', 'execute') does not match the actual tool. No bundled scripts to test. The skill is essentially an elaborate fiction documenting a CLI that doesn't exist as described.
Major accuracy issue: the skill documents a CLI tool that doesn't match reality. The real 'browser-use' is a Python AI agent library for browser automation, not a CLI with session/cookie/execute subcommands. This means agents following these instructions would hit command-not-found errors or install the wrong package. Security is fine — no malicious patterns. Architecture is weak: monolithic SKILL.md with no scripts, no references/, no separation. Usefulness is low because the documented CLI doesn't exist; agents can't actually use these commands.
