- CREATION
- REFINEMENT
- MODIFICATION
- HUMAN
Short answer
How can I make and edit images using my AI?
You provide text descriptions to create new images and then use specific commands to swap backgrounds or add elements to a scene. You'll get high-fidelity visuals and precise control over the final composition.
Visual outcomes
What you'll create
The different ways your AI handles visual requests.
CREATION
New image assets
The AI builds images from scratch based on your text. You get unique, high-fidelity files ready for use.
REFINEMENT
Cleaned subjects
The AI strips away distracting backgrounds. You get a clean subject isolated for your next design step.
MODIFICATION
Expanded scenes
You can add objects or change parts of an existing image. The AI fills in the gaps to match the original style.
HUMAN
Creative direction
You decide which versions look right and provide the specific descriptive tweaks. The AI follows your lead on style.
The workflow
What your AI does when the request arrives.
The AI handles the heavy lifting of rendering and pixel manipulation.
Generate the base image
Your AI takes your descriptive prompt and renders a high-fidelity image from nothing.
text_to_imageIsolate the subject
If the image has a busy background, the AI can strip it away to leave just the main object.
remove_backgroundModify the scene
You can ask to add a hat to a character or change the lighting in a specific area.
generative_fillExpand the frame
The AI uses generative logic to extend the edges of a photo or fix small errors.
generative_fill
Try it
Copy these to start.
Use these prompts to trigger the image tools immediately.
Starting points
These are opening lines. Swap out the bracketed text with your own specific descriptions.
Adobe Firefly Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_oON4wNSLZkI3u30Uts0iISrmeHToWjdps809izQk/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Adobe Firefly capabilities are ready to use.
{
"mcpServers": {
"adobe-firefly-mcp": {
"url": "https://edge.vinkius.com/vk_preview_oON4wNSLZkI3u30Uts0iISrmeHToWjdps809izQk/mcp"
}
}
}Create a text_to_image of a futuristic cyberpunk city with neon lights and heavy rain.
Use remove_background on the image I just generated to isolate the central character.
Apply generative_fill to this image to add a vintage leather jacket to the person.
Use generative_fill to change the background of this photo to a sunny tropical beach.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Adobe Firefly once, then ask.
Link your account once. Your login stays encrypted on our side, and the full walkthrough lives on the Connector page.
Connect Adobe Firefly to your AIFAQ
How this task behaves
- 01
Can the AI edit existing photos I upload?
Yes, once you provide the image, the AI can use generative tools to modify it.
- 02
Can the AI create video from my text?
No, the current tools are limited to generating and editing still images.
- 03
Will the AI automatically save my images to my hard drive?
No, the AI generates the images within the chat interface. You must manually save the files you want to keep.
- 04
Can the AI remove specific objects from a photo?
Yes, using the generative fill capability, you can instruct the AI to replace or remove elements.
- 05
Does the AI handle the actual graphic design layout?
No, the AI handles the image generation and pixel edits. You are responsible for the final layout and design composition.
More questions about Adobe Firefly? The Connector page answers them. See everything the Adobe Firefly Connector can do
Connect Adobe Firefly to Claude, Cursor, ChatGPT & more
