Make images and video from Claude with MCP

MCP is the standard way for an AI assistant to use tools. Connect the studio and your agent can list the models, start an image or a video, and check when it is done, with the same models, prices and credits as you have in the studio.

You decide what it may do: the connection asks for your permission, and every run shows its price before it starts.

The main ones, named the way your agent sees them.

• list_models, to find the right model for the job.
• generate_image and generate_video, to start a run.
• get_execution_status, to know when it is ready.
• build_influencer_prompts, save_character and get_character, to build a character and use it again.

A writing agent that makes its own blog image for every article. Claude Code that makes the screenshots and covers for a site while it builds it. A pipeline that builds a character once and keeps putting it in new scenes.