The Agent
The Kyoso agent is your AI creative partner — it reasons, plans, and executes multi-step design work across images, video, and animation.
What it is
The Kyoso agent is the core of the product — an AI creative partner that understands your intent, plans how to get there, and executes. You describe what you need in plain language, and the agent reasons about the best approach, selects the right tools and models, and generates or edits images and video for you.
Today you interact with the agent through the agent input on the board. You can guide it with model picks, brand context, and selected references — or just describe what you want and let it decide.
Key ideas
- Reasoning and planning — the agent can plan multi-step workflows, suggest creative directions, describe and analyze images, run multiple actions in sequence, and remember your brand guidelines throughout the session
- Selection-aware — the agent reads what you have selected on the board; selected nodes are treated as references that the agent can analyze and work with
- Brand knowledge — when working inside a brand, the agent draws on trained brand guidelines, assets, and styles to keep every generation on-brand
- Quick actions — selecting nodes surfaces applicable tools so you don't have to remember what's available; you can also select nodes and ask the agent directly
@mentions — type@in the agent input to pick a model, an aspect ratio, an uploaded logo, a font, or other assets- Generation queue — you can queue up to 5 generations in a single agent session
How it fits with the rest of Kyoso
The agent is where all of Kyoso's tools, image models, and video models come together. Tools give the agent specialized capabilities, models give it creative range, and your brand gives it visual identity. The board is the shared workspace where you see the agent's output and guide its next steps.