AI and AgentsSimplified view
How an MCP request flows
A walk through one Model Context Protocol request, from the client through authentication, memory and the model to a server that coordinates tools.
- Model Context Protocol
- OAuth
- LLM tool calling
8 stepsPress play, or use next and previous to follow the flow one step at a time.
All steps
What this shows
The Model Context Protocol (MCP) gives AI clients a standard way to reach tools and data. This diagram follows a single request: the client authenticates, loads memory, involves a model, and then hands the work to an MCP server that routes it to tools.
Design notes
- One protocol, many tools. The server presents each tool the same way, so adding a tool does not change the client.
- Auth first. Identity is settled before memory, models or tools are touched.
- The tools shown are examples. Search, code analysis and a database query stand in for whatever a server exposes.
For a longer explanation, read the MCP server guide.
More workflows
Test Automation
NL2Test, from a sentence to a Playwright test
How NL2Test turns a plain-language description into an executable Playwright test, with a model planning the test and an MCP server driving a real page.
10 steps 9 parts Simplified view
AI and Agents
Multi-agent supervisor
A supervisor agent receives a request, hands sub-tasks to specialist agents one at a time, keeps shared memory, and combines their results into one answer.
8 steps 7 parts Simplified view
AI and Agents
RAG pipeline
How retrieval-augmented generation works end to end, from loading documents into a vector store to answering a question with the closest passages.
8 steps 7 parts Simplified view