Information-centric networking assumes that content is the focus of communication and is uniquely named in networks e.g. a consumer asks a network for specific content instead of connecting to the IP address of a specific server. This approach allows content to be cached within the network and network nodes to change their location without affecting the availability of content. Current implementations of ICN approaches such as NDN provide proof-of-concept solutions that demonstrate the viability of information-centric networking.
Model Context Protocol (MCP) servers provide a connection between AI agents and existing data sources and servers. Interactions with MCP servers provide AI agents with the ability to query and modify state outside an AI workflow. For example, an MCP server connected to a Kubernetes cluster could allow users of Claude or ChatGPT to query the state of deployments and adjust the details of the cluster by talking to the AI implementation e.g. the query “How many pods have been restarted in the last 10 minutes?” would be translated by the AI implementation into a query to the MCP server which in turn would know how to retrieve the information from the Kubernetes cluster.
The aim of this project is to investigate how a MCP server can be applied to deployments of information-centric networking in order to query and control the state of the deployment.