Skip to content

Lesson 3 of 6

Store context in a Project

Create a Project, add your improved prompt as project instructions, and upload the text file so repeated requests use the same background.

5 minClaude, Gemini and ChatGPT

After this lesson, you will be able to create a Claude Project, add project instructions, and upload a text file so that repeated requests use the same background.

What is a Project?

A Project in Claude is a workspace where you can store instructions and files that apply to every conversation inside it. Instead of pasting the same prompt and the same text into each new chat, you set them once. Claude then uses that context for all your requests in the Project.

You will use the prompt you improved in the previous lesson. If you did not save it, create a file called prompt.txt with this content:

You are a patient teacher who explains technical ideas to beginners.

Explain the text below in simple words.

Write three bullet points. Each bullet point must be one sentence.

Text:

You will also need the Pandas text from the first lesson. Save it in a file called pandas.txt:

Pandas is a Python library for working with data. It provides two main data structures: Series for one-dimensional data and DataFrame for two-dimensional data. You can use Pandas to read data from CSV files, clean it, and analyse it.

Create a Project

Open Claude chat in your web browser. On the left side, find the Projects section and click the button to create a new Project. Give it the name Summariser and a short description such as Summarise technical texts for beginners.

After you create the Project, you see a page with two areas: Project instructions and Project knowledge. Project instructions are text that Claude follows in every conversation in this Project. Project knowledge is a set of files that Claude can read when needed.

Add project instructions

Click in the Project instructions area and paste the content of prompt.txt. Do not include the Text: line at the end; you will add the text as a file instead. The instructions should look like this:

You are a patient teacher who explains technical ideas to beginners.

Explain the text below in simple words.

Write three bullet points. Each bullet point must be one sentence.

These instructions tell Claude what role to take and what output format to use. Because they are in the Project, you do not need to repeat them in each message.

Upload a text file

In the Project knowledge area, click the button to upload a file. Choose pandas.txt from your computer. Claude will process the file and make its content available to the Project.

Now start a new conversation inside the Project. Type a short request that refers to the file, such as:

Summarise the pandas.txt file.

Claude uses the project instructions and the uploaded file to produce the summary. The answer should be three bullet points, each one sentence, written in simple words.

Use the Project for repeated requests

You can start another conversation in the same Project and ask a different question about the same file. For example:

What is a DataFrame?

Claude still has the file and the instructions, so it can answer based on the text you uploaded. You do not need to paste the text again.

To change the behaviour for all conversations, edit the project instructions. For example, change the output format to Write a single paragraph of no more than 50 words. Then start a new conversation and ask the same summarise request. The new answer follows the updated instruction.

Projects are useful when you have a set of files or rules that you use often. Instead of copying them every time, you put them in a Project and let Claude handle the context.

Key takeaways

  • A Project stores instructions and files that apply to every conversation inside it.
  • Project instructions define the role and output format for all requests in the Project.
  • Uploaded files become Project knowledge that Claude can use without you pasting their content.
  • You can edit project instructions at any time; new conversations use the updated version.
  • Projects save time when you work repeatedly with the same background material.

Try it

Create a new Project called Glossary. Add project instructions that tell Claude to act as a technical editor and to define any technical term it finds in the uploaded text. Upload a short text file of your choice that contains at least two technical terms. Start a conversation and ask Claude to list the terms and their definitions. Check that the answer includes every technical term from your file and that each definition is clear. A good result is a list that you could use as a glossary for a colleague.

Official documentation

Ready to use this in a job?

Rolebound applies to matching roles for you.

Start now

Written with AI and checked automatically. Tools change, so a command may behave differently in the version you install. Spotted a mistake? Tell us.

    Cookies on Rolebound

    We use necessary cookies to sign you in and keep the site working — those cannot be turned off. We would also like to set analytics and advertising cookies so we can see which pages help people get interviews and which adverts brought them here. Nothing optional is set unless you say yes, and you can change your mind at any time from “Cookie settings” in the footer. Read our Privacy Policy.

    Store context in a Project | Working with Claude: Prompting, Projects and API | Rolebound