browser-site-skill

Convert exploratory browser automation results into reusable site-skill documents.

Updated Aug 31, 2024
One-click install
npx skills add https://github.com/iheCoder/Lib --skill browser-site-skill
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: browser-site-skill
Source: https://github.com/iheCoder/Lib/tree/main/skill/browser-site-skill
Command: npx skills add https://github.com/iheCoder/Lib --skill browser-site-skill

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Browser automation often relies on brittle selectors; this skill provides a stable approach by exploring tasks and distilling successful flows into reusable site skills.

Core Features & Use Cases

  • Two modes: exploration/execution and deposition into site skills.
  • Multimodal anchors including semantic_name, visible_text, aria_role, dom_path_summary, and state_before/state_after with validation.
  • Drift signals and fallback strategies to handle site changes and data sources.
  • Template-driven output enabling quick creation of site-skill documents.

Quick Start

Run a browser task on a target site and emit a reusable site skill document from the successful workflow.

Frequently Asked Questions about browser-site-skill

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

FAQPage Schema
How do I stabilize browser automation when site selectors keep changing?

Stabilize browser automation by converting exploratory task results into reusable site skills, using multimodal anchors like semantic names and aria roles instead of brittle selectors. This approach validates state changes and outputs a structured skill document for reuse.

What is the best way to automate web testing workflows across similar websites?

Automate web testing workflows by exploring a target site, validating execution steps, and depositing the successful flow into a reusable site-skill document. This structured output includes drift signals and fallback data to handle site changes across similar websites.

How does multimodal anchoring work for browser workflow automation?

Multimodal anchoring for browser workflow automation uses semantic names, visible text, aria roles, and DOM path summaries to identify elements. It validates state before and after actions, ensuring stable execution even when specific site attributes change.

Can I reuse browser automation flows after a website updates its layout?

Yes, you can reuse automation flows after website updates because the site skill document includes drift signals and fallback strategies. These mechanisms detect layout changes and automatically switch to alternative data sources or anchors to maintain execution stability.

Why does my browser automation break during site exploration and execution?

Browser automation breaks during site exploration because it relies on brittle DOM selectors. This skill solves the problem by validating before and after states, using multimodal anchors, and generating a deposition document with fallback data for unstable elements.

Do I need specific dependencies to generate site skills for web automation?

No specific dependencies are required to generate site skills for web automation. The skill operates independently to explore tasks, validate states, and output a template-driven site-skill document with drift signals and fallback data for workflow reuse.