Abstract
In enterprise video conferencing environments, capturing actionable engineering artifacts such as bug reports, architecture diagrams, or task assignments often requires users to manually context-switch across multiple browser tabs and applications. Existing third-party artificial intelligence (AI) agents attempt to address this by ingesting meeting transcripts via backend APIs, but this approach introduces security risks by transmitting confidential meeting data to external servers and requiring highly-privileged OAuth scopes for cross-application execution. The present disclosure describes a client-side autonomous orchestrator that operates entirely within the user's authenticated browser session. The orchestrator utilizes front-end robotic process automation and Document Object Model (DOM) manipulation to pilot pre-authenticated AI panels, inject constrained prompts, monitor virtual DOM layout shifts for generation completion, extract structured data, and dynamically generate pre-filled URI payloads for cross-application routing. All artifacts are persisted locally in browser storage. This approach enables real-time execution of multi-application AI workflows without transmitting meeting data to external servers.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sawant, Mansi, "Client-Side DOM-Based Orchestration of Generative AI Meeting Artifacts", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/11703