browser_action

The browser_action utility enables Sypha to interact directly with web-resident applications and sites. Utilizing a managed headless browser, Sypha can navigate pages, perform click events, and input text, while providing visual validation through automated screenshots.

Parameters

• action (required): Select from launch, click, type, scroll_down, scroll_up, or close.
• url (optional): The target web destination (required for the launch operation).
• coordinate (optional): Specific x,y screen coordinates for targeting clicks.
• text (optional): The specific string to be entered into a form or input field.

Workflow Integration

1. Initialization (Launch): Sypha establishes a fresh browser session at the target URL.
2. Sequential Interaction: Sypha processes instructions one by one. Following each interaction, it captures a high-resolution screenshot and persists session logs for analysis.
3. Session Termination (Close): Once the objective is reached, the browser must be explicitly closed to release system memory and resources.

Strategic Applications

• Automated UI Validation: "Verify that the primary navigation bar is visible and responsive on the landing page."
• Internal Environment Auditing: "Access the staging dashboard and report the current status of the service health-check."
• Operational Automation: "Log in to the cloud console and trigger a manual database snapshot."

Environment Modes

• Managed Instance (Default): Sypha launches a fresh, isolated Chromium instance for the session.
• Remote Bridge: Connects to a pre-existing Chrome process (ideal for testing sites with active login sessions or specialized security configurations).

[!IMPORTANT] During an active browser session, Sypha's focus is dedicated to web-interaction. You must close the browser session before Sypha can resume utilizing standard file system or terminal utilities.