browserbase-create

0
0
Source

Guide Claude through creating new browser automation scripts using the stagehand CLI

Install

mkdir -p .claude/skills/browserbase-create && curl -L -o skill.zip "https://mcp.directory/api/skills/download/9054" && unzip -o skill.zip -d .claude/skills/browserbase-create && rm skill.zip

Installs to .claude/skills/browserbase-create

About this skill

Create Automation Skill

Guide Claude through creating new browser automation scripts using the stagehand CLI.

When to Use

Use this skill when:

  • User wants to automate a website task
  • User needs to scrape data from a site
  • User wants to create a Browserbase Function
  • Starting from scratch on a new automation

Workflow

1. Understand the Goal

Ask clarifying questions:

  • What website/URL are you automating?
  • What's the end goal (extract data, submit forms, monitor changes)?
  • Does it require authentication?
  • Should this run on a schedule or on-demand?

2. Explore the Site Interactively

Start a local browser session to understand the site structure:

stagehand session create --local
stagehand goto https://example.com

Use snapshot to understand the DOM:

stagehand snapshot

Take screenshots to see the visual layout:

stagehand screenshot -o exploration.png

3. Identify Key Elements

For each step of the automation, identify:

  • Selectors for interactive elements
  • Wait conditions needed
  • Data to extract

Use the accessibility tree refs to understand element relationships:

[@0-5] button: "Submit"
[@0-6] textbox: "Email"
[@0-7] textbox: "Password"

4. Test Interactions Manually

Before writing code, verify each step works:

stagehand fill @0-6 "[email protected]"
stagehand fill @0-7 "password123"
stagehand click @0-5
stagehand wait networkidle
stagehand snapshot

5. Enable Network Capture (if needed)

For API-based automations or debugging:

stagehand network on
# perform actions
stagehand network list
stagehand network show 0

6. Create the Function

Once you understand the flow, create a full function project:

stagehand fn init my-automation
cd my-automation

This creates a complete project with:

  • package.json with dependencies
  • .env with credentials (from ~/.stagehand/config.json if available)
  • tsconfig.json
  • index.ts template

Edit index.ts with your automation logic:

import { defineFn } from "@browserbasehq/sdk-functions";
import { chromium } from "playwright-core";

defineFn("my-automation", async (context) => {
  const { session } = context;
  const browser = await chromium.connectOverCDP(session.connectUrl);
  const page = browser.contexts()[0]!.pages()[0]!;

  // Your automation steps here
  await page.goto("https://example.com");
  await page.fill('input[name="email"]', context.params.email);
  await page.click('button[type="submit"]');
  
  // Extract and return data
  const result = await page.textContent('.result');
  return { success: true, result };
});

7. Test Locally

Start the local development server:

pnpm bb dev index.ts
# or: stagehand fn dev index.ts

Then invoke locally via curl:

curl -X POST http://127.0.0.1:14113/v1/functions/my-automation/invoke \
  -H "Content-Type: application/json" \
  -d '{"params": {"email": "[email protected]"}}'

8. Deploy to Browserbase

When ready for production:

pnpm bb publish index.ts
# or: stagehand fn publish index.ts

9. Test Production

Invoke the deployed function:

stagehand fn invoke <function-id> -p '{"email": "[email protected]"}'

Best Practices

Selectors

  • Prefer data attributes (data-testid) over CSS classes
  • Use text content as fallback (text=Submit)
  • Avoid fragile selectors like nth-child

Waiting

  • Always wait for navigation/network after clicks
  • Use waitForSelector for dynamic content
  • Set reasonable timeouts

Error Handling

  • Wrap risky operations in try/catch
  • Return structured error information
  • Log intermediate steps for debugging

Data Extraction

  • Use page.evaluate() for complex extraction
  • Validate extracted data before returning
  • Handle missing elements gracefully

Example: E-commerce Price Monitor

defineFn("price-monitor", async (context) => {
  const { session, params } = context;
  const browser = await chromium.connectOverCDP(session.connectUrl);
  const page = browser.contexts()[0]!.pages()[0]!;

  await page.goto(params.productUrl);
  await page.waitForSelector('.price');

  const price = await page.evaluate(() => {
    const el = document.querySelector('.price');
    return el?.textContent?.replace(/[^0-9.]/g, '');
  });

  return {
    url: params.productUrl,
    price: parseFloat(price || '0'),
    timestamp: new Date().toISOString(),
  };
});

You might also like

flutter-development

aj-geddes

Build beautiful cross-platform mobile apps with Flutter and Dart. Covers widgets, state management with Provider/BLoC, navigation, API integration, and material design.

1,4071,302

drawio-diagrams-enhanced

jgtolentino

Create professional draw.io (diagrams.net) diagrams in XML format (.drawio files) with integrated PMP/PMBOK methodologies, extensive visual asset libraries, and industry-standard professional templates. Use this skill when users ask to create flowcharts, swimlane diagrams, cross-functional flowcharts, org charts, network diagrams, UML diagrams, BPMN, project management diagrams (WBS, Gantt, PERT, RACI), risk matrices, stakeholder maps, or any other visual diagram in draw.io format. This skill includes access to custom shape libraries for icons, clipart, and professional symbols.

1,2201,024

ui-ux-pro-max

nextlevelbuilder

"UI/UX design intelligence. 50 styles, 21 palettes, 50 font pairings, 20 charts, 8 stacks (React, Next.js, Vue, Svelte, SwiftUI, React Native, Flutter, Tailwind). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app, .html, .tsx, .vue, .svelte. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient."

9001,013

godot

bfollington

This skill should be used when working on Godot Engine projects. It provides specialized knowledge of Godot's file formats (.gd, .tscn, .tres), architecture patterns (component-based, signal-driven, resource-based), common pitfalls, validation tools, code templates, and CLI workflows. The `godot` command is available for running the game, validating scripts, importing resources, and exporting builds. Use this skill for tasks involving Godot game development, debugging scene/resource files, implementing game systems, or creating new Godot components.

958658

nano-banana-pro

garg-aayush

Generate and edit images using Google's Nano Banana Pro (Gemini 3 Pro Image) API. Use when the user asks to generate, create, edit, modify, change, alter, or update images. Also use when user references an existing image file and asks to modify it in any way (e.g., "modify this image", "change the background", "replace X with Y"). Supports both text-to-image generation and image-to-image editing with configurable resolution (1K default, 2K, or 4K for high resolution). DO NOT read the image file first - use this skill directly with the --input-image parameter.

970608

pdf-to-markdown

aliceisjustplaying

Convert entire PDF documents to clean, structured Markdown for full context loading. Use this skill when the user wants to extract ALL text from a PDF into context (not grep/search), when discussing or analyzing PDF content in full, when the user mentions "load the whole PDF", "bring the PDF into context", "read the entire PDF", or when partial extraction/grepping would miss important context. This is the preferred method for PDF text extraction over page-by-page or grep approaches.

1,033496

Stay ahead of the MCP ecosystem

Get weekly updates on new skills and servers.