skills.rest
kkudumu
kkudumuCOMMUNITY
Explore all skills by this author →
GitHub

ftm-browse

Manage a persistent headless Chromium daemon for browser automation via local API.

Updated Aug 23, 2026
Search Tags:#automation#cli#playwright#browser-daemon#headless-browser#visual-verification#aria-tree
One-click install
npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-browse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: ftm-browse
Source: https://github.com/kkudumu/feed-the-machine/tree/main/ftm-browse
Command: npx skills add https://github.com/kkudumu/feed-the-machine --skill ftm-browse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a persistent, on-demand headless browser daemon to perform visual verification, form filling, and web interaction via a CLI/API, enabling agents to automate and verify frontend tasks without manual browser sessions.

Core Features & Use Cases

  • Persistent headless Chromium daemon accessible via a CLI binary and HTTP API
  • Snapshot ARIA trees, locate interactive elements, and perform actions like goto, click, fill, screenshot
  • Integration with other FTM skills for visual verification, UI regression checks, and automated workflows

Quick Start

Start the daemon and navigate to a page, then take a screenshot to verify rendering.

Frequently Asked Questions about ftm-browse

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

FAQPage Schema
How do I automate visual verification of web pages using a headless browser?▼

Visual verification is automated by running a persistent headless Chromium daemon that takes screenshots and inspects ARIA trees. This allows agents to visit pages and verify frontend rendering without manual browser sessions.

What is a persistent browser daemon and why use one for automated UI testing?▼

A persistent browser daemon is a local server managing a headless Chromium instance accessible over an API. It enables agents to maintain session state and perform sequential web interactions without repeatedly launching the browser.

How do I inspect and interact with ARIA trees in an automated browser workflow?▼

You inspect ARIA trees by sending requests to the daemon's local API to snapshot the page structure. The daemon locates interactive elements, enabling automated actions like clicking, filling forms, and evaluating page state across sessions.

Can I use Playwright for headless browser automation with a CLI and API?▼

Yes, this Skill requires a Playwright-based browser engine to power the headless Chromium daemon. It exposes a local token-based API and a CLI client that automatically spins up the daemon when needed for automation tasks.

How do I fill forms and click elements in a headless browser using an API?▼

Form filling and element clicking are executed by sending commands to the local daemon API. The daemon navigates to the target page, locates the interactive elements via ARIA snapshots, and performs the specified actions over a persistent session.

Do I need to manually start the browser daemon before running automation tasks?▼

No, the CLI client automatically spins up the local daemon server when needed. It manages the persistent headless Chromium instance and supports optional persistent user data directories for supervised execution.

Related Skills

Explore More Software Engineering Skills→
lyh2lyh2

gstack

Automate QA browser sessions with a deterministic headless Chromium environment.

Community
Advanced
viably-labsviably-labs

playwright-screenshot-inspector

Analyze Playwright screenshots to validate UI rendering accuracy semantically.

Official
Advanced
everrunseverruns
42

ui-screenshots

Capture web page screenshots with agent-browser for visual verification.

Official
Intermediate
fajshahfajshah
1

browsing-with-playwright

Automate browser navigation, form interactions, and data extraction via Playwright MCP.

Community
Intermediate