Flashcard

You hear about 'eval harnesses' like Promptfoo and DeepEval. Explain what an eval harness does that a Jupyter notebook cannot.

Before shipping a prompt change, the lead says 'run it against the golden set.' What is a golden dataset and why does every team need one?

Where do you configure MCP servers in Zed, and where do their tools appear?

You want to build an MCP server in TypeScript. What package do you use and how does it handle schemas?

Walk me through what tools/list returns and when the host calls it.

Your MCP tool just failed. How does the server tell the model what went wrong?

You want to build an MCP server in Python. What do you install and what does the API look like?

In MCP, how is a notification different from a request?

In MCP, what exactly is a 'tool' and who decides when to call it?

Your teammate says they built an 'MCP server'. What did they actually build?

MCP has tools, resources, and prompts. What makes a resource different from a tool?

Why does MCP have a 'prompt' primitive separate from tools and resources?

Your MCP server builds but the host can't see its tools. What's the first debugging move?

How is an MCP host different from an MCP client?

Someone asks you what MCP is at a meetup. How do you explain it?

Where does the MCP client live, and why is it one to one with a server?

A local stdio MCP server has no auth token. Is that a security hole?

Why did MCP pick JSON-RPC 2.0 as its wire format instead of REST or gRPC?

GitHub ships an official MCP server. What can it do, and how does it authenticate?

The filesystem MCP server gives agents file access. How does it stop them from reading everything on disk?

How does Cursor discover and load MCP servers?

You need to add an MCP server to Claude Desktop. Which file do you edit and what goes in it?

What happens during the MCP initialize handshake, and why does it matter?

A tokenizer has a 'vocabulary'. What is it, and why does its size matter for the model?

Your team debates 32K vs 128K vs 256K vocab. What is the core tradeoff they should frame?