lens-inspect

Name: lens-inspect
Availability: InStock
Author: sendbird-playground

Inspect live web apps via MCP tools for DOM, console, and network data.

3|1|Updated Mar 25, 2026

Search Tags:#automation #debugging #mcp #browser #lens #dom

Ready to run

npx skills add https://github.com/sendbird-playground/stave --skill lens-inspect

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: lens-inspect
Source: https://github.com/sendbird-playground/stave/tree/main/skills/lens-inspect
Command: npx skills add https://github.com/sendbird-playground/stave --skill lens-inspect

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Lens Inspect provides AI agents with a real-time browser in Stave to inspect, interact with, and debug live web applications by exposing DOM, console, and network data in-context.

Core Features & Use Cases

Embedded Lens browser for live inspection of pages, DOM access, console and network readouts
MCP toolset to navigate, screenshot, get_html, get_text, evaluate JavaScript, click elements, type, and snapshot accessibility
Use cases include UI debugging, design-spec verification, and locating source code references quickly

Quick Start

Open Lens from the right rail and enter a URL to start inspecting the page.

Frequently Asked Questions about lens-inspect

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

FAQPage Schema

How do I inspect DOM and console data from a live web app for debugging?▼

You can inspect a live web app by opening the embedded Lens browser in Stave, navigating to a URL, and using the MCP toolset to read DOM, capture console logs, and monitor network data for real-time UI debugging.

Can I take screenshots and click elements during an automated browser inspection?▼

Yes, automated browser inspection supports taking screenshots and clicking elements. The MCP toolset includes click, type, and screenshot functions, allowing you to interact with live pages and capture visual states for verification.

Does the Lens browser require a workspaceId to run MCP inspection commands?▼

Yes, a workspaceId is required to operate the Lens inspection tool. This identifier enables the MCP-enabled workflow, allowing you to execute commands like get_html, evaluate JavaScript, and snapshot accessibility within your Stave environment.

What is the best way to verify UI design specs against a running application?▼

The best way to verify UI design specs is to use the Lens browser in Stave to navigate your live app, take screenshots, and read DOM text. This allows direct comparison of rendered elements against your design requirements.

Why does my automated web inspection fail to capture network requests?▼

Automated web inspection may fail if the workspaceId is missing or the Lens browser is not properly opened from the Stave right rail. Ensure the MCP toolset is connected to execute the get_network command and collect request data.