Manus released "Design View", an AI design workspace within Manus Agent, focusing on merging "idea-drawing-refinement" into the same canvas process. The core update is the Mark Tool: users can directly circle the area on the graph and write down instructions to achieve more granular local modifications, and try to avoid repeatedly rewriting prompts or regenerating the entire image for small changes.
The official example shows that Mark Tool can issue modification requirements for multiple objects in the same image, and complete the replacement and adjustment while maintaining the original light, shadow, composition and style as much as possible. At the same time, it provides the ability to "edit text", which can extract text elements from the screen and modify them. The mobile terminal also supports long-press circle selection and one-time batch submission of multiple changes. Manus said that this model is open to all users, but the local editing tools are more effective on images generated within the platform, and the actual output still needs to pay attention to copyright and commercial compliance risks.
FAQs
Q: What problems does Manus Design View mainly solve?
A: It is used to modify the "last mile" details of the generated image to make it controllable, reducing the repeated change of prompt words and regeneration of the whole image.
Q: How exactly does Mark Tool work?
A: Circle the target area on the picture, and then enter a simple modification command to issue different requirements for multiple areas and process them in batches.
Q: Does the mobile terminal support the same circle modification?
A: Yes, the mobile terminal can be edited by long-pressing the circled area and using voice or text input commands.
Q: Can I also refine the uploaded external image?
A: You can upload them for typography and inspiration, but local editing tools are usually more stable on images generated within Manus.
