Lab Task 6: Agentic and orchestration layers
Agentic AI communication protocols serve as the foundational layer enabling what researchers and industry leaders envision as the “Internet of Agents” – a distributed ecosystem where autonomous AI agents can discover each other’s capabilities, establish trust relationships, and collaborate to solve complex real-world problems.
The implications of this technological shift extend far beyond technical considerations, touching on fundamental questions of security, governance, and the future of human-AI collaboration.
Building infrastructure for the Internet of Agents
The AGNTCY project provides the complete infrastructure stack for agent collaboration—discovery, identity, messaging, and observability that works across any vendor or framework. It is the foundational layer that lets specialized agents find each other, verify capabilities, and work together on complex problem.
MCP 101
Traditional application programming interfaces (APIs) have long served as the backbone of software integration, providing deterministic, well-defined interfaces for system-to-system communication. However, the unique characteristics of Agentic AI systems is including their autonomous decision-making capabilities, natural language processing requirements, and dynamic behavioral adaptation. This has exposed significant limitations in conventional API-based communication models.
The Model Context Protocol, introduced by Anthropic in late 2024, focuses primarily on standardizing how AI agents access external tools and data sources, providing a universal interface that connects agents to diverse resources without requiring custom integrations for each data silo.
Until now, individual integrations had to be developed for different AI systems, even when accessing the
same data sources and tools.
Data sources and tools are available on MCP servers. The AI systems access them through a unified mechanism, which prevents redundant integrations.
Step 1: Login to LibreChat
Login to LibreChat with your credentials.
Step 2: Choose the large language model
Choose the large language model based on the screenshot below:
Step 3: Enable the MCP server
In the following example, we are focusing on NetBox and Catalyst Center:
Step 4: Enter the prompts
Remember to replace the pod ID in the prompt!
Show me all information about POD01-CAT8KV-01 from Catalyst Center and Netbox