Beyond the vocabulary: what these features actually do
Plain-language explainers for four features people ask about most, each summarised from Anthropic's own documentation with a link to the source. Product details change often, so treat the linked page as the current word on it — this page is a starting point, not the last word.
Projects
A self-contained workspace with its own chat history and a knowledge base you build up — upload documents, add standing instructions, and every chat inside that Project can draw on them. Useful for keeping one AI "workspace" per class, per client, or per research thread instead of re-explaining context every time. Free accounts can create up to five; paid plans get more room and larger knowledge bases.
Artifacts
When a response is substantial and self-contained — a code file, a document, a small web page, a diagram — Claude can open it in its own side panel instead of burying it in the chat. You can keep iterating on it, and some artifacts can be published so others can open and use them directly.
Connectors & MCP ("Plugins")
Connectors let Claude read from or act on outside tools — calendars, email, project trackers, and more — through an open protocol called MCP (Model Context Protocol), which Anthropic created and published openly. Anthropic curates a directory of reviewed connectors, but anyone can also add a custom connector that meets the same technical spec without waiting for directory approval.
Claude Code
An agentic coding tool that works directly against a real codebase — reading files, making edits, running tests and commands, and handling Git — rather than just answering questions about code in a chat window. It runs from a terminal, an IDE extension, a desktop app, or the browser, and can be scripted or scheduled for repeating jobs like PR reviews.
Common questions
Short answers, each sourced to Anthropic's help center. Click a question to expand it.
Where to verify anything on this page
This lab summarises and interprets; it doesn't set policy or document features. These are the primary sources — go here for anything you need to cite or rely on.
Official product documentation
- anthropic.comCompany site — products, news, policy
- docs.claude.comDeveloper & API documentation
- support.claude.comClaude.ai help center — features, limits, how-tos
- code.claude.com/docsClaude Code documentation
Research & safety publications
- anthropic.com/researchAnthropic's research index
- Constitutional AI (arXiv 2212.08073)Bai et al., 2022 — the paper behind Claude's alignment method
- anthropic.com/newsModel releases, system cards, policy updates