International Journal of Science, Engineering and TechnologyAuthors: Partha Sarathi Samal, Suresh Kumar Palus, Sai Kiran Padmam, Bhavan Kumar B R
Abstract: Media supply chains currently suffer from severe fragmentation. Tools for transcoding, quality control (QC), and media asset management (MAM) exist in isolated silos. Connecting these systems requires custom, brittle API integrations that fail when vendor specifications change. AI Agents offer a solution through autonomous planning and execution. However, agents struggle to connect to diverse legacy systems without a standardized interface. This paper proposes a unified framework based on the Model Context Protocol (MCP). MCP acts as a universal transport layer. It allows AI agents to discover, query, and manipulate media assets securely across heterogeneous environments. We outline a "Media-MCP-Agent" architecture that enables autonomous agents to parse technical metadata, trigger transcode jobs and rectify QC failures without hard-coded scripts. We provide a rigorous analysis of the security implications, specifically regarding prompt injection, and discuss the future of hierarchical agent swarms in broadcast operations.
DOI: http://doi.org/10.5281/zenodo.18104044
