SECAI Core · phase 5 of 15
Model context protocol
AI systems, agents, and tools to share data about memory, goals, and intermediate results, enabling multi-agent systems, orchestration frameworks, or AI assistants to coordinate seamlessly
The Explain card
- Plain English
- Model Context Protocol (MCP) is a standard way for AI systems, agents and tools to share context such as memory, goals and intermediate results, so assistants and multi-agent frameworks can coordinate.
- Example
- A coding assistant connects to an MCP server that exposes the file system and a database. The assistant can now list tables and read files on request, which is powerful, and also means a compromised MCP server can feed poisoned data or over-broad capabilities straight into the agent.
- Why it matters
- MCP turns models into actors with tools. Defenders need to inventory MCP servers, restrict their permissions, and treat their outputs as untrusted input.
- Hook
- MCP is the USB port for AI agents: handy, standard, and worth checking before you plug anything in.
Where it sits in the deck
Phase 5: Interacting with Models: Prompting, APIs, and Retrieval
With a working model and clean data, learn the interaction layer — how users and systems communicate instructions and retrieve grounded answers.
