skills.rest
akillness
akillnessCOMMUNITY
Explore all skills by this author →
Website𝕏@akillness0308GitHub

browser-harness

Connect LLM agents to Chrome via the Chrome DevTools Protocol for autonomous browser automation.

40|9|Updated Jan 4, 2026
Search Tags:#llm#autonomy#cdp#chrome#self-healing#browser-automation#domain-skills
One-click install
npx skills add https://github.com/akillness/jeo-skills --skill browser-harness-akillness
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: browser-harness
Source: https://github.com/akillness/jeo-skills/tree/main/.agent-skills/browser-harness
Command: npx skills add https://github.com/akillness/jeo-skills --skill browser-harness-akillness

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Autonomous browser control for language models, enabling direct browser interaction via Chrome DevTools Protocol to run tasks without manual UI operations.

Core Features & Use Cases

  • Direct CDP-based browser control for LLM agents to inspect pages, form interactions, and verify results.
  • Self-healing helpers and domain-skills to extend site-specific behavior across platforms.
  • Cross-platform compatibility with Claude Code, Codex CLI, Gemini CLI, and OpenCode.

Quick Start

Clone the skill repository, create and activate a Python 3.10+ virtual environment, and run the browser harness to connect to a locally debuggable Chrome instance.

Frequently Asked Questions about browser-harness

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

FAQPage Schema
How do I enable LLM browser automation using Chrome DevTools Protocol?▼

You can enable LLM browser automation by connecting this skill to a locally debuggable Chrome instance via CDP. It requires Python 3.10+ and provides self-healing helpers to handle login flows, navigation, and data extraction autonomously.

What is self-healing browser automation for AI agents?▼

Self-healing browser automation allows LLM agents to dynamically adapt to page changes during tasks like form filling and navigation. Domain-skills extend site-specific behavior, ensuring autonomous workflows remain stable across platforms like Claude Code, Codex CLI, Gemini CLI, and OpenCode.

Do I need a specific Python environment to run CDP browser automation?▼

Yes, autonomous browser control requires Python 3.10+ and a Chrome or Chromium instance with remote debugging enabled. You also need a local workspace containing agent_helpers.py and domain-skills files for extending site-specific behavior.

Can I use autonomous browser control with Gemini CLI and OpenCode?▼

Yes, autonomous browser control supports Gemini CLI, OpenCode, Claude Code, and Codex CLI. The skill uses CDP to connect your LLM agent to Chrome, enabling cross-platform tasks like login flows, data extraction, and verification.

How do I extend site-specific browser behavior for LLM agents?▼

You extend site-specific browser behavior by customizing the domain-skills and agent_helpers.py files in your local workspace. This allows the LLM agent to handle unique page structures and self-heal during autonomous navigation and form interactions.

Why does my LLM agent fail to control Chrome via CDP?▼

LLM agents fail to control Chrome via CDP if the browser is not launched with remote debugging enabled or if the Python environment is below 3.10. Ensure Chrome is running locally with the correct debugging port and the workspace contains the required helper scripts.

Related Skills

Explore More Software Engineering Skills→
GraffiohGraffioh

web-browser

Automate Chrome browser interactions via the Chrome DevTools Protocol.

Community
Advanced
smarmen86smarmen86

blog-factcheck

Validate statistical claims in blog posts against cited source URLs.

Community
Intermediate
MH4GFMH4GF
3

difit

View Git diffs in a browser and generate AI review prompts.

Community
Intermediate
nbiishnbiish

skyvern

Automate browser workflows, form filling, and data extraction using self-hosted Skyvern with Playwright.

Community
Advanced