What problem does it solve?
AI agents lack the ability to read visual debugging interfaces like React DevTools or the Next.js dev overlay, yet they need detailed component trees, props, hooks, PPR shell states, build errors, and network traffic to effectively debug Next.js applications. This Skill converts every interactive panel a developer would click through into structured text output that an LLM can parse, reason about, and act upon in a loop.
Core Features & Use Cases
- React Component Tree Inspection: Retrieve the full component hierarchy using the tree command, then drill into any node to see props, hooks, state, and source-mapped file locations.
- PPR Shell Analysis: Use ppr lock and ppr unlock to freeze and analyze Partial Prerendering shells, identifying which Suspense boundaries create dynamic holes and what blocks them.
- Error and Network Diagnostics: View build errors, runtime exceptions, dev server logs, and full network request details including headers and response bodies.
- Visual and Runtime Inspection: Capture screenshots, control viewport size, evaluate JavaScript in page context, and read the Next.js error overlay from shadow DOM.
Quick Start
Use the next-browser skill to open your Next.js dev server and inspect the React component tree to find rendering bottlenecks.