browser-mcp-explore

Attach to live browser sessions and inspect pages for debugging.

44|8|Updated Jul 14, 2025
One-click install
npx skills add https://github.com/thegeronimo/hyperopen --skill browser-mcp-explore
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-mcp-explore
Source: https://github.com/thegeronimo/hyperopen/tree/main/.agents/skills/browser-mcp-explore
Command: npx skills add https://github.com/thegeronimo/hyperopen --skill browser-mcp-explore

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The Browser MCP Explore skill enables rapid live browser debugging, session inspection, and parity checks to validate flows before committing Playwright coverage.

Core Features & Use Cases

  • Live browser attach and inspection of an existing tab for immediate visibility into states and selectors.
  • Exploratory, flaky, or diagnostic tasks that benefit from governed review and artifact-heavy evidence.
  • Parity comparisons and design reviews to surface reproducible steps before automation is formalized.
  • Selector or flow discovery to surface exact repro steps before transforming paths into Playwright coverage.

Quick Start

Attach to a running browser session and start exploratory inspection with the Browser MCP workflow to surface selectors, states, and repro steps.

Frequently Asked Questions about browser-mcp-explore

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I debug flaky UI tests by inspecting a live browser session?

Debug flaky UI tests by attaching to live browser sessions to inspect current page states, discover exact selectors, and capture reproducible paths for evidence. The skill connects to existing tabs to provide immediate visibility into the specific conditions causing test failures.

What is the best way to find CSS selectors before writing Playwright automation?

Finding CSS selectors for Playwright automation is done through live browser inspection and selector discovery. You attach to a running tab to interact with the page, identify exact element selectors, and trace the reproducible steps needed to formalize your test coverage paths.

Can I use a live browser attach for design review and parity checks?

Live browser attach supports design review and parity checks by allowing direct inspection of running pages to validate UI states. You can compare visual flows against design specifications, surface discrepancies, and collect artifact-heavy evidence before committing to automation.

Do I need an existing MCP server registration to inspect browser flows?

An existing MCP server registration at /hyperopen/.codex/config.toml is required to inspect browser flows. The skill relies on this configuration to attach to live sessions and stores all collected debugging artifacts under /hyperopen/tmp/browser-inspection/ for traceable evidence.

How do I collect debugging artifacts during exploratory browser testing?

Collect debugging artifacts during exploratory browser testing by using the live MCP inspection workflow, which automatically stores evidence under /hyperopen/tmp/browser-inspection/. This enables traceable debugging by saving page states, discovered selectors, and reproducible paths encountered during the session.

Why does my Playwright test coverage fail without reproducible browser paths?

Playwright test coverage fails without reproducible paths because automation cannot target unstable selectors or unpredictable states. Live browser inspection surfaces exact repro steps and selector states before formalizing paths, ensuring the resulting test scripts execute reliably against validated flows.