Working with Images

Upload images to your project, embed them in documents, or let the agent generate and process them — all without leaving Obvious.

Upload an Image

Drag and drop an image file into the chat or click the + button next to the message bar and select your file. Obvious accepts JPEG, PNG, GIF, and WebP formats up to 500 MB.

Uploaded images appear in your project's file library.

Use Images in Documents

Embed any image in a document using markdown syntax:

![Description of the image](image-url)

The agent handles this for you. Drop an image into the chat and ask:

Add this image to my document.

For banner images at the top of a document, open the document's ⋯ menu and select Add cover image, or ask the agent:

Add a cover image of a product dashboard to this document.

Generate Images with AI

Ask the agent to create an image from a description:

Generate a flat-style illustration of a team collaborating around a data dashboard, blue and purple tones.

The agent generates the image and saves it to your project files. Embed it in a document, use it as a cover image, or download it.

Specific prompts get better results — include style, colors, composition, and mood. For anything with text or labels (diagrams, wireframes, mockups), ask for "pro quality" to get sharper rendering.

To edit a generated image, reference it in a follow-up message:

Make the background darker and add a subtle grid pattern.

The agent modifies the existing image rather than starting from scratch.

Let the Agent Process Your Images

The agent can read and analyze images you upload. Drop a screenshot, chart, or photo into the chat and ask:

What data is shown in this chart?

Extract the text from this whiteboard photo.

The agent sees the image directly — no extra steps needed.

Next Steps

• Cover Images — add a banner to the top of a document.
• Uploading and Managing Files — manage all file types in your project.