Working with files: Uploads and Context library

Your Context library is your context hub in Juma. It stores all your files, images, web pages, and instructions in one place so you can reuse them across different chats and Projects without uploading them again.

How to access your Context library

You can open your Context library in two ways:

• Click the plus button in the chat input field
• Use the @ shortcut in the chat field

What you can store in your Context library

Your library can different types of content: instructions, web pages, files and images.

Instructions

Instructions let you guide how the AI responds. For example, you might create an instruction that says "Write in a conversational, friendly tone. Use short sentences and avoid jargon." Once you save it, you can reuse it every time you start a new chat.

To create a new instruction:

• Click Add context in your library
• Select Instructions
• Enter your instruction text
• Name your instruction
• Click Save

Example instructions for marketers:

• "Always write headlines under 60 characters and use active voice"
• "Format all social media captions with emojis and include 3-5 relevant hashtags"
• "Write all blog content at an 8th grade reading level and break up text with subheadings every 200 words"

Web pages

You can save website content as context. To add a web page:

• Go to the Web Pages tab in your library
• Paste your URL
• Click Save

Files and images

You can upload files and images in two ways. Drag and drop them directly into a chat, or click Add context in your library and choose Files or Images. Either way, your uploads will be processed and stored permanently in your library.

Supported file types:
pdf, doc, docx, xls, xlsx, pptx, txt, htm, html, json, csv, md, cpp, hpp, c (max 25 MB)

Supported image types:
jpg, jpeg, png, webp, svg, gif, avif (max 30 MB)

Integrations

You can connect files directly from Google Drive, HubSpot, Notion, OneDrive, and SharePoint. This lets you pull in content from these platforms without manually uploading it.

Using context in your chats

To add context to a chat, open your Context library and select the items you need. Selected items appear as tags in the chat input field.

You can hover over any context item to preview its contents. If something isn't relevant for your current chat, you can remove it by clicking the X on the tag.

Context in Projects

When you're working inside a Project, you'll see all the context that's attached to that Project's knowledge. If you want to add additional files or instructions, you can either drag and drop them into the chat or open your Context library and select what you need.

Privacy and organization

Your Context library is private. Everything you upload stays secure and organized so you can reuse it across different chats and Projects, whether they're private or shared with your team.