Browser Automation Resource

Note: This capability is now provided as an installable component. See the Components guide for how to install and use it.

Install: kdeps component install browser

Usage: run: { component: { name: browser, with: { url: "...", action: "navigate", selector: "...", screenshotPath: "..." } } }

The Browser component enables browser automation via Playwright, supporting page navigation, text extraction, and screenshots.

Note: The component exposes three actions: navigate, screenshot, and getText. For advanced automation (form fill, click, evaluate JavaScript, sessions, stealth mode), use a Python resource with playwright directly.

Component Inputs

Input	Type	Required	Default	Description
url	string	yes	—	URL to navigate to
action	string	no	navigate	Action: navigate, screenshot, or getText
selector	string	no	—	CSS selector for getText action
screenshotPath	string	no	/tmp/screenshot.png	File path for screenshot action output

Using the Browser Component

Navigate and get page text:

run:
  component:
    name: browser
    with:
      url: "https://example.com"
      action: getText

Take a screenshot:

run:
  component:
    name: browser
    with:
      url: "https://example.com"
      action: screenshot
      screenshotPath: "/tmp/page.png"

Get text from a specific element:

run:
  component:
    name: browser
    with:
      url: "https://example.com"
      action: getText
      selector: ".article-content"

Access the result via output('<callerActionId>').

---

Actions

Action	Description
navigate	Navigate to the URL and return the final page URL
screenshot	Capture a PNG screenshot of the page
getText	Extract visible text from the page or from a CSS selector

---

Result Map

Key	Type	Description
success	bool	true if the action completed without error.
text	string	Extracted text (getText action).
screenshotPath	string	Path to saved screenshot (screenshot action).
url	string	Final page URL after navigation.

---

Expression Support

All fields support KDeps expressions:

run:
  component:
    name: browser
    with:
      url: "{{ get('target_url') }}"
      action: getText
      selector: "{{ get('css_selector') }}"

---

Full Example: Screenshot Pipeline

# Step 1: Take a screenshot of a dashboard
- apiVersion: kdeps.io/v1
  kind: Resource

  metadata:
    actionId: capture
    name: Capture Dashboard

  run:
    component:
      name: browser
      with:
        url: "{{ get('dashboard_url') }}"
        action: screenshot
        screenshotPath: /tmp/dashboard.png

# Step 2: Send the screenshot by email
- apiVersion: kdeps.io/v1
  kind: Resource

  metadata:
    actionId: notify
    name: Send Screenshot
    requires:
      - capture

  run:
    component:
      name: email
      with:
        to: "{{ get('recipient') }}"
        subject: "Dashboard snapshot"
        body: "Screenshot saved at: {{ output('capture').screenshotPath }}"
        smtpHost: "{{ env('SMTP_HOST') }}"
        smtpUser: "{{ env('SMTP_USER') }}"
        smtpPass: "{{ env('SMTP_PASS') }}"

---

Installation

Playwright browsers must be installed on the host running kdeps:

npx playwright install chromium
# or install all browsers:
npx playwright install

---

Related Resources

• HTTP Client - Simple HTTP requests without a browser
• Scraper - Text extraction from already-fetched pages
• Python - Custom Python automation scripts (for advanced Playwright use)