An AI Agent for Auditing Frontend Applications
summary
An approach to building an AI agent that audits frontend applications by analyzing code, styles and runtime metrics (Lighthouse, DevTools, and custom analyzers). It explains the underlying tools and protocols for such an agent, how it adapts to different projects, and how it converts disparate data into actionable insights to augment — not replace — the development team.
AI-written summary · about
chapters (11)
- 0:00Frontend audit with AI
- 7:04What is frontend auditing
- 13:24Problems with audit tools
- 17:40Automating audits with CDP and ACore
- 23:36Integrating tools into AI agent
- 27:30Collecting user scenarios
- 31:01Optimizing large traces
- 33:52CSS audit and performance
- 36:02Security and SEO audits
- 38:50Orchestrating AI audit agent
- 40:58Demo and conclusions
related talks
23mHuman-Centered AI: Supporting, Not Replacing, Frontend Developers
35mThe Agentic AI Engineer
14mAI for Web3 Security: Real-World Use Cases — Dr. Stefan Beyer
17mAI-Driven Frontend Development: Beyond Traditional
26mAI Agents as Your Co-Developer
32mLeveraging Angular-Based Frameworks for AI-Driven Testing
01h 28mAI Coding Agents and How to Code Them
53mDid AI Kill Frontend?