Plain English
Claude Code Glossary
The words you'll hear around Claude Code and AI building - MCP servers, agents, skills, hooks, tokens, and more - explained simply, no jargon.
- Claude Code
- Anthropic's agentic coding tool that runs in your terminal (and in editors and the desktop/web app). You describe what you want in plain English and Claude reads, writes, and edits real files in your project, runs commands, and iterates with you - so you can build software without hand-writing every line.
- Claude
- The family of AI models built by Anthropic that powers Claude Code. Different versions (such as the Opus, Sonnet, and Haiku tiers) trade off speed, cost, and capability. Claude Code uses Claude under the hood to understand your requests and produce code.
- Agent
- An AI that can take a goal, decide on the steps to reach it, and carry them out using tools - reading files, running commands, calling APIs - with little step-by-step instruction. Claude Code is itself an agent; you can also build your own agents with it.
- Subagent
- A separate, focused agent that the main agent can hand a specific task to - for example, exploring a codebase or running tests - so the work happens in its own context. Subagents let you parallelize work and keep the main conversation clean.
- MCP (Model Context Protocol)
- An open standard that lets AI tools like Claude Code connect to external data and services in a consistent way. Think of it as a universal adapter: instead of custom one-off integrations, MCP gives the model a standard way to talk to tools.
- MCP Server
- A small program that exposes a specific tool or data source (a database, your file system, GitHub, Slack, a browser, etc.) to Claude Code over the Model Context Protocol. Add an MCP server and Claude can use that capability directly inside your workflow.
- Skill
- A reusable, packaged capability you can give Claude Code so it knows how to perform a specific task the way you want - your conventions, steps, and tools bundled together. Skills turn 'explain it every time' into 'invoke it once'.
- Slash command
- A shortcut, typed as /name, that expands into a longer instruction or workflow inside Claude Code. Slash commands save you from retyping common prompts and keep your team's workflows consistent.
- Hook
- A script you configure to run automatically in response to events in Claude Code - for example, before or after a tool runs. Hooks let you enforce rules, run checks, or trigger actions without asking the model to remember to do them.
- CLAUDE.md
- A project file where you write standing instructions and context for Claude Code - your stack, conventions, and preferences. Claude reads it on each turn, so you write things down once instead of re-explaining yourself in every prompt.
- Context window
- The amount of text (measured in tokens) an AI model can consider at one time - your prompts, files, and the conversation so far. When work exceeds the window, older context must be summarized or dropped, which is why focused tasks and good context management matter.
- Token
- The unit AI models use to measure text - roughly a word-piece. Prompts and responses are counted in tokens, and both pricing and the context window are measured in them.
- Prompt
- The instruction you give the AI. In Claude Code, a good prompt is specific and concrete - it names the outcome you want - which is the single biggest lever on the quality of what you get back.
- System prompt
- The behind-the-scenes instructions that set an AI's role, rules, and behavior before your messages begin. It shapes how the model responds across the whole conversation.
- Tool use
- When an AI model calls an external function or service - running a command, editing a file, searching the web - rather than only producing text. Tool use is what turns a chatbot into an agent that can actually do things.
- Vibe coding
- Building software by describing what you want to an AI and iterating on the results, rather than writing the code by hand. It's a fast way to get to a working product - and it works best when paired with reviewing what the AI produces.
- Prompt engineering
- The practice of writing and structuring prompts to reliably get the output you want - being specific, giving context and examples, and breaking big asks into steps. With agentic tools it's less about clever wording and more about clear direction.
Stop Googling the jargon. Start building.
We'll teach you every one of these hands-on. Join Claude Code Club for $9/month.
Related: curriculum, guides, and the blog.