Skip to main content
All CollectionsCorti AssistantIntegrations
Integrating Corti Assistant via iFrame or Webview
Integrating Corti Assistant via iFrame or Webview

Learn how to embed Corti Assistant into another application using iFrames or WebViews.

Updated over 2 months ago

Corti Assistant can be embedded into third-party platforms using either an iFrame or a webview. This allows partners to integrate Corti’s real-time AI-powered assistance and automated documentation into their workflows without requiring users to switch between applications.

Integrating with iFrame

An iFrame allows embedding one webpage inside another. This option is for customers who are developing a browser-based application.

Recommended iFrame Configuration

<iframe name="iframe" id="frame" title="Corti Assistant" src="{webAppUrl}" sandbox="allow-forms allow-modals allow-pointer-lock allow-popups allow-popups-to-escape-sandbox allow-same-origin allow-scripts allow-downloads" allow="microphone *; camera *; device-capture *"></iframe>

Note: Since iFrames require manual permission settings, future updates may modify these recommendations as Corti Assistant expands functionality.

Integrating with WebView

Corti Assistant supports embedding via webviews on modern platforms, with the following compatibility:

  • Fully Supported Webviews:

    • Edge WebView2 for Windows

    • Android System WebView

    • Electron

  • Partially Supported: (certain web features are not supported in Safari-based browsers)

    • WebKit on OSX

    • WKWebView on iOS

Generating the Web App URL

The web app URL you pass to your iFrame or WebView depends on your environment, but it is typically one of the following:

To give end users the most seamless experience, it is advised to log the user in server side first and pass that token to the web application through query parameters.

Passing Credentials to the Web App

Partners must authenticate users securely to access Corti Assistant. The recommended authentication flow follows these steps:

  1. Authenticate using OAuth2.

  2. Take the full auth response and Base64 encode it.

    await response.text();
    const b64_auth_response = btoa(data);
  3. Append the encoded auth response as a query parameter to your launch URL:

    e.g. https://assistant.corti.app?auth_response=<b64>

This ensures a secure and streamlined login process.

Embedded Mode

Partners can choose between two integration approaches:

  1. Full Navigation: Direct the user to the homepage, allowing full access to Corti Assistant.

  2. Restricted Mode: Lock navigation to a specific feature using the embedded=true query parameter. This keeps the interface in kiosk mode, hiding side navigation and ensuring users remain within a controlled workflow.

Is an API Required for iFrame/Webview Integration?

Basic embedding does not require an API. However, API access is necessary for deeper integration:

  • Retrieving user-specific data (e.g., fetching patient records from an EHR system).

  • Saving Corti-generated summaries into external platforms (e.g., CRM, EHR).

  • Implementing advanced authentication beyond basic login.

Did this answer your question?