Skip to content

JPeetz/agent-skills / skills/qa-testing/browser-automation

browser-automation

Verified

Use this skill when automating browser interactions with Playwright — testing, scraping, monitoring, form submission, screenshot capture, and multi-page interaction flows. Covers setup, navigation, element interaction, assertions, and cleanup. Designed for agentic platforms — Claude Code, Codex, Cursor, Gemini CLI, OpenClaw, GitHub Copilot, Windsurf, and OpenCode.

FileKindSize
CHANGELOG.mdmarkdown1.4 KB
LICENSEother1.0 KB
SKILL.mdskill md23.2 KB
evals/evals.jsonconfig4.9 KB
references/browser-testing-workflows.mdmarkdown23.1 KB
references/playwright-patterns.mdmarkdown12.1 KB
references/selector-strategies.mdmarkdown10.3 KB
scripts/playwright_setup.pyexecutablescript8.1 KB

evals/evals.json

View on GitHub
{
  "skill": "browser-automation",
  "version": "1.0.0",
  "cases": [
    {
      "id": "pos-01-form-submission",
      "description": "Explicit request for automated browser interaction \u2014 form filling, clicking, and assertion.",
      "input": "Write a test that fills out the registration form at https://example.com/signup with name, email, and password, clicks submit, and verifies it redirects to the welcome page.",
      "shouldTrigger": true,
      "expect": "navigate to /signup / fill name, email, password fields / click submit button / assert redirect to /welcome",
      "assertions": [
        "navigate to /signup",
        "fill name, email, password fields",
        "click submit button",
        "assert redirect to /welcome"
      ]
    },
    {
      "id": "pos-02-screenshot-capture",
      "description": "Direct request for automated screenshot capture via browser automation.",
      "input": "Take a full-page screenshot of the pricing page at https://example.com/pricing and save it as pricing.png.",
      "shouldTrigger": true,
      "expect": "navigate to /pricing / capture full_page screenshot / save to pricing.png",
      "assertions": [
        "navigate to /pricing",
        "capture full_page screenshot",
        "save to pricing.png"
      ]
    },
    {
      "id": "pos-03-element-visibility",
      "description": "E2E testing request with specific element visibility checks and content assertions.",
      "input": "Write an e2e test that checks the search bar is visible on the homepage, types 'laptop' into it, and asserts that at least 5 results appear.",
      "shouldTrigger": true,
      "expect": "navigate to homepage / assert search bar visible / type 'laptop' into search / trigger search / assert at least 5 results displayed",
      "assertions": [
        "navigate to homepage",
        "assert search bar visible",
        "type 'laptop' into search",
        "trigger search",
        "assert at least 5 results displayed"
      ]
    },
    {
      "id": "pos-04-data-extraction",
      "description": "Classic web scraping request with structured data extraction.",
      "input": "Scrape all product names and prices from https://books.toscrape.com and output them as JSON.",
      "shouldTrigger": true,
      "expect": "navigate to the page / locate product elements / extract name and price for each / output as JSON",
      "assertions": [
        "navigate to the page",
        "locate product elements",
        "extract name and price for each",
        "output as JSON"
      ]
    },
    {
      "id": "pos-05-multi-page-flow",
      "description": "Synthetic monitoring of a multi-page user flow with error handling.",
      "input": "Create a monitoring script that goes through our checkout flow: add item to cart \u2192 view cart \u2192 enter shipping \u2192 enter payment \u2192 confirm order. Alert if any step fails.",
      "shouldTrigger": true,
      "expect": "multi-step navigation through checkout pages / form filling at each step / error detection per step / alert on failure",
      "assertions": [
        "multi-step navigation through checkout pages",
        "form filling at each step",
        "error detection per step",
        "alert on failure"
      ]
    },
    {
      "id": "neg-01-browser-feature-question",
      "description": "Asking about browser features/capabilities \u2014 no automation intent. This is a factual question, not a request to automate browser interaction.",
      "input": "What browsers support the WebGPU API, and what's the current adoption rate?",
      "shouldTrigger": false,
      "expect": "",
      "assertions": [
        "Asking about browser features/capabilities \u2014 no automation intent. This is a factual question, not a request to automate browser interaction."
      ]
    },
    {
      "id": "neg-02-playwright-api-question",
      "description": "Asking for Playwright API documentation/explanation \u2014 no concrete automation task. This is an educational question about the tool, not a request to use it.",
      "input": "How does page.waitForSelector work in Playwright? Can you explain the different wait states?",
      "shouldTrigger": false,
      "expect": "",
      "assertions": [
        "Asking for Playwright API documentation/explanation \u2014 no concrete automation task. This is an educational question about the tool, not a request to use it."
      ]
    },
    {
      "id": "neg-03-manual-testing-request",
      "description": "Requesting manual visual review/design feedback \u2014 not automated testing. The user wants a human opinion on aesthetics, not programmatic verification.",
      "input": "Can you open Chrome and check if the new landing page looks good? I want feedback on the hero section design.",
      "shouldTrigger": false,
      "expect": "",
      "assertions": [
        "Requesting manual visual review/design feedback \u2014 not automated testing. The user wants a human opinion on aesthetics, not programmatic verification."
      ]
    }
  ]
}
finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 322kUpdated Claude CodeCodexVerifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 322kUpdated Claude CodeCodexVerifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 322kUpdated Claude CodeCodexVerifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 198kUpdated Claude CodeCodexVerifiedInstall