Abstract
Enterprises increasingly rely on meetings as the primary unit of collaboration for sensitive decision-making, training, customer engagement, and incident response. However, current firewall and network security systems remain largely static and topology-driven, forcing administrators to manually create, deploy, and later remove temporary security policies for short-lived events. This approach is operationally complex, error-prone, and does not scale in modern hybrid and highly collaborative environments.
The proposal introduces a meeting-centric, agent-driven firewall policy framework that automatically derives security intent from meeting context and participant composition. By treating the meeting identifier as the authoritative security signal, the system dynamically forms a meeting-specific identity group, selects a pre-approved security policy template, and enforces it as a time-bounded policy overlay for the duration of the meeting. Policies are automatically reverted at meeting completion, eliminating the risk of lingering temporary rules.
The framework operates independently of physical location, SSIDs, or static network segmentation, making it suitable for single-SSID wireless deployments, hybrid meetings, and mobile users. It enables proactive, intent-aware security enforcement without relying on deep packet inspection or reactive traffic analysis. Built-in monitoring and assurance ensure correctness, automatic rollback, and full auditability, providing clear traceability from business intent to enforcement.
By aligning security controls with what is happening and who is involved, rather than where traffic originates, the proposal closes a critical gap in existing firewall automation. It reduces human error, improves compliance, and introduces a new class of intent-aware capabilities that differentiate modern firewall platforms.
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
M M, Niranjan, "AN AGENTIC AI SYSTEM FOR MEETING-CENTRIC, TIME-BOUNDED SECURITY POLICY ENFORCEMENT", Technical Disclosure Commons, (May 18, 2026)
https://www.tdcommons.org/dpubs_series/10154