browser-agent-debug

Debug browser automation failures using CDP-connected Playwright sessions on port 9222.

1|Updated Apr 11, 2026
One-click install
npx skills add https://github.com/drgon1/santian --skill browser-agent-debug
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-agent-debug
Source: https://github.com/drgon1/santian/tree/main/ai-skills/skills/browser-agent-debug
Command: npx skills add https://github.com/drgon1/santian --skill browser-agent-debug

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps debug browser automation workflows that fail in AI coding tools, especially when login state, selectors, popups, or page transitions cause unreliable execution.

Core Features & Use Cases

  • Screenshot-guided debugging: Inspect the page before and after every action so you can verify state changes instead of guessing.
  • CDP-connected browser control: Reuse an existing logged-in browser session through a fixed remote debugging port for consistent testing.
  • Cross-tool compatibility: Work with multiple AI programming environments while controlling the same persistent browser instance.
  • Practical use cases: Repair publish flows, fix broken selectors, diagnose iframe editors, validate cover uploads, and troubleshoot login or popup issues in web automation scripts.

Quick Start

Ask the AI to connect to the existing browser session, capture the current page state with a screenshot, and guide the next debugging step one action at a time.

Frequently Asked Questions about browser-agent-debug

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

FAQPage Schema
How do I debug browser automation failures caused by broken selectors or popups?

Debug browser automation failures by capturing screenshots before and after actions to verify state changes. This stepwise inspection repairs broken selectors, handles popups, and validates page transitions instead of guessing execution states.

Can I reuse an existing logged-in browser session for Playwright automation?

Yes, you can reuse an existing logged-in browser session by connecting via CDP on port 9222. This maintains login state and enables consistent Playwright-based interaction across multiple AI programming environments without re-authenticating.

What is the best way to repair a web publishing flow that fails during automated execution?

Repair a failing web publishing flow through CDP-connected browser control and screenshot-driven debugging. Diagnose iframe editors, validate cover uploads, and troubleshoot login issues one action at a time for reliable execution.

Does this screenshot debugging approach work with my current AI coding tools?

Yes, this screenshot debugging approach works across multiple AI coding tools. It controls the same persistent browser instance through a fixed remote debugging port, ensuring cross-tool compatibility while maintaining stepwise verification.

Why does my browser automation script lose its login state between test runs?

Browser automation scripts lose login state when launching isolated instances. Reuse logged-in browser contexts through a CDP-connected session on port 9222 to maintain persistent authentication and ensure consistent execution.