What you will be able to do
- Decide when a one-off chat is enough and when recurring work belongs in a Project
- Explain what a Project's knowledge base and instructions do, and what they do not share between chats
- Identify which Project capabilities depend on plan tier (project limits, RAG, sharing)
- Recognise when Claude's output should be an artifact and what must be enabled for artifacts to work
Key concept
Project (workspace with persistent context) — A Project is a separate workspace. Documents you upload and instructions you set there apply to every chat you start inside it. It is the tool for recurring work that needs the same background each time, not for a single question.
1.The first choice: a plain chat or a Project
Before you choose a model or turn on a tool, decide where the work should happen. A plain chat suits a self-contained task: you ask, Claude answers, and the task ends. Recurring work is different. Think of a budgeting spreadsheet you revisit every week, a client account with a standing brief, or a codebase you keep asking about. Pasting the same background into every new conversation is wasted effort, and a Project exists to remove it.
The help centre describes a Project as a self-contained workspace with its own chat histories and its own knowledge base. In each one you upload documents, add context, and hold focused chats with Claude. That makes a Project an organising tool as well: conversations about one topic stay together, apart from unrelated chats about recipes or travel.
Many people assume this is a paid feature, but it isn't. Projects are available to every user, including those on free accounts. The only restriction on the free tier is a cap on how many Projects you can create.
Sources1
2.What a Project carries into every chat, and what it doesn't
A Project gives Claude two kinds of persistent context. The first is the project knowledge base, which sits on the right side of the project page. Anything you upload there (documents, text, code) is used across all of the chats in that Project. The second is project instructions, set through "Set project instructions". They tailor how Claude responds in every chat in the Project, for example with a more formal tone or from the viewpoint of a particular role or industry. The name and description you give the Project are different: Claude cannot see them.
Not by default. Chats in a Project share the knowledge base and the instructions, not each other's conversations. If Chat B needs the decision you explained in Chat A, add it to the project knowledge base. Only content saved there reaches every chat.
What about a Project whose knowledge grows past what fits in the context window? On paid plans (Pro, Max, Team, Enterprise), Claude switches to Retrieval Augmented Generation (RAG) mode automatically when project knowledge approaches the context limit. It then uses a project knowledge search tool to pull the relevant parts instead of loading everything at once. This raises capacity by up to 10x, and you don't have to set anything up. If the knowledge later drops back below the threshold, Claude can return to in-context processing on its own. Before RAG, a Project could not accept more content once it reached the context-window threshold.
| Capability | Free | Pro / Max | Team / Enterprise |
|---|---|---|---|
| Create Projects with knowledge base and instructions | Yes, up to five | Yes | Yes |
| Enhanced project knowledge with RAG | No | Yes | Yes |
| Share a Project with organisation members | No | No | Yes |
Projects are not a way to save quota. They use your plan's allowance like any other use of Claude.
3.One shared place for a team's context
Suppose several colleagues each upload the same reference documents into their own separate chats. The feature built for this is a shared Project. On Team and Enterprise plans, a Project can be shared with other members of the organisation. Several members can then add documents, create chats and work in the same Project environment.
The owner shares by opening the Project and clicking "Share project". They can add people one at a time by name or email, or paste in a list of addresses. Each person gets one of two permission levels. Can view lets a member see the contents, knowledge and instructions and chat within the Project, but not edit it. Can edit also lets them change the instructions and knowledge and update member settings. A Project can also be made visible to the whole organisation, either when it is created or later.
Admins can override all of this. Owners can switch off project sharing for the organisation, and on Enterprise they can do so for specific roles. Turning off sharing also turns off public projects. Existing shares remain in place, but nothing new can be shared. If a colleague can't share a Project, check the organisation settings before assuming the feature doesn't exist.
A legal team repeatedly asks Claude to draft memos that reference the same set of contract templates and a shared style guide. They want every new conversation to automatically have access to those documents without re-uploading them each time. Which Claude product feature best fits this need?
Correct answer: A — Create a Project and upload the contract templates and style guide to its project knowledge
- A. Correct. Projects maintain a persistent knowledge base that Claude references across all conversations within that project, so uploading the templates and style guide once makes them available to every future memo without re-uploading.
- B. Pasting the same documents into every new conversation defeats the purpose of persistent context and requires manual repetition each time.
- C. Research mode performs agentic web and connected-source searches; it does not give Claude access to private documents that were never uploaded or connected.
- D. An artifact is a standalone content window for a single conversation's output, not a mechanism for supplying persistent background knowledge across future conversations.
Sources1
4.Artifacts: when the output is the deliverable
Projects decide where context lives. Artifacts decide what happens to the output. The help centre describes an artifact as anything Claude makes for you that you'd put in front of someone: a design, a deck, a document, a dashboard or a small interactive tool. It opens in its own window beside the conversation, where you can keep editing it, come back to it and share it. Everything you make is saved to the Artifacts tab in the sidebar, so you can find it again from any conversation.
Claude produces an artifact when the content meets these criteria: it is significant and self-contained (typically over 15 lines), you're likely to edit, iterate on or reuse it outside the conversation, it stands on its own without the chat around it, and you're likely to refer back to it. A short factual answer doesn't qualify. A playable single-page prototype you plan to tweak all afternoon does. You can change an artifact by asking Claude or, for template-based ones, by editing it directly, and it updates live. To try a different direction without losing your current work, edit an earlier message. That creates a new version of the chat with its own artifacts.
| Feature | Free | Pro, Max, Team, Enterprise |
|---|---|---|
| Create artifacts in a chat | Yes | Yes |
| Start from a template (Design, Slides, Docs) | No | Yes (beta; off on Enterprise until an owner turns each on) |
| Connect your apps to an artifact | No | Yes |
| Store data in an artifact | No | Yes |
Artifacts can also call Claude directly, which turns them into small apps. Anyone using a shared app signs in with their own Claude account, and the usage counts against their plan limits, not the creator's. The platform also differs: on iOS and Android you can ask for an artifact and view it, but starting from a template, editing and changing sharing settings require the web or Desktop app.
A developer asks Claude to build a small interactive React component that visualizes sales data with adjustable filters, and expects to iterate on the code across several follow-up messages. Which feature will display this output in a way that supports iterative editing and version history?
Correct answer: A — The artifact panel, which shows the component separately and tracks versions as it is revised
- A. Correct. Artifacts open substantial, self-contained content like a React component in a dedicated side panel and provide a version selector, letting the developer iterate and switch between revisions.
- B. Project knowledge is meant for background reference documents the user uploads, not for displaying and iteratively editing generated code with version tracking.
- C. Research mode is for gathering external information through web searches, not for rendering or iterating on generated code.
- D. Printing code inline in the main transcript does not provide the dedicated side-panel view or version history that iterative component development benefits from.
Sources4
Exam traps
Each one states something that sounds right. Open it to see what is actually true.
1.Projects are a paid feature, so a free user has to upgrade to keep a topic's conversations in a dedicated space.Why is that wrong?
Every plan, including Free, has Projects. The free tier is only capped at five of them.
Covered in The first choice: a plain chat or a Project
2.Chats in the same Project automatically remember each other's conversations.Why is that wrong?
Chats in a Project share the knowledge base and instructions, not each other's history. Anything every chat needs must go into the project knowledge base.
Covered in What a Project carries into every chat, and what it doesn't
3.Artifacts are always available and need no settings.Why is that wrong?
Artifacts depend on the Cloud code execution and file creation capability. If an admin or user has turned it off, no artifacts are created.
Sources
Every claim above is drawn from one of these pages, quoted as it was written on the date shown.
- 1.
“Projects allow you to create self-contained workspaces with their own chat histories and knowledge bases.”
↩︎ The first choice: a plain chat or a Project“Free users can create a maximum of five projects.”
↩︎ The first choice: a plain chat or a Project“When your project knowledge approaches context limits, Claude seamlessly enables RAG mode to expand capacity by up to 10x while maintaining response quality.”
↩︎ What a Project carries into every chat, and what it doesn't“Projects draw from your plan's usage like any other use of Claude”
↩︎ What a Project carries into every chat, and what it doesn't“For users on Team and Enterprise plans, projects can be shared with other members of your organization”
↩︎ One shared place for a team's context“Can view: Members can see project contents, knowledge, and instructions, and chat within the project, but cannot edit it.”
↩︎ One shared place for a team's context“When sharing is off, users can't share projects with new users or groups, and existing shares stay in place.”
↩︎ One shared place for a team's context“Projects allow you to create self-contained workspaces with their own chat histories and knowledge bases.”
↩︎ Key concept“Projects are available to all users, including those with free Claude accounts. Free users can create a maximum of five projects.”
↩︎ Exam trap 1 - 2.
“Anything you upload to this space will be used across all of your chats within that project.”
↩︎ What a Project carries into every chat, and what it doesn't“Context is not shared across chats within a project unless the information is added into the project knowledge base.”
↩︎ What a Project carries into every chat, and what it doesn't“Give your project a name and description (note that Claude will not have access to these details).”
↩︎ What a Project carries into every chat, and what it doesn't“Context is not shared across chats within a project unless the information is added into the project knowledge base.”
↩︎ Exam trap 2 - 3.https://support.claude.com/en/articles/11473015-retrieval-augmented-generation-rag-for-projectsOfficial docs
“If your project knowledge later drops below the context window threshold, Claude can automatically convert back to context-based processing.”
↩︎ What a Project carries into every chat, and what it doesn't - 4.https://support.claude.com/en/articles/17153992-what-are-artifacts-and-how-do-i-use-themOfficial docs
“It opens beside your conversation, and you can edit it, come back to it, and share it with others.”
↩︎ Artifacts: when the output is the deliverable“It's something you're likely to want to edit, iterate on, or reuse outside the conversation.”
↩︎ Artifacts: when the output is the deliverable“Artifacts require Cloud code execution and file creation to be turned on in Settings > Capabilities”
↩︎ Artifacts: when the output is the deliverable“Edit an earlier message to create a different version of the chat, with its own artifacts”
↩︎ Artifacts: when the output is the deliverable“usage counts against each person's own plan limits rather than yours”
↩︎ Artifacts: when the output is the deliverable“Artifacts require Cloud code execution and file creation to be turned on in Settings > Capabilities”
↩︎ Exam trap 3