New GoogleSlidesTools toolkit

We’ve introduced GoogleSlidesTools to give agents full control over Google Slides. With it, you can create presentations, build out slides, and manage content end to end, all directly from your agent.

Agents can add and reorder slides, insert text boxes, tables, images, and videos, and read existing slide content to stay context-aware. Whether you are building decks from scratch or modifying existing ones, everything happens programmatically in a single workflow.

We support both OAuth and service account authentication, so you can use the toolkit in interactive setups or deploy it in server-side, multi-user environments.

from agno.agent import Agent
from agno.models.google import Gemini
from agno.tools.google.slides import GoogleSlidesTools

agent = Agent(
    model=Gemini(id="gemini-2.0-flash"),
    tools=[
        GoogleSlidesTools(
            oauth_port=8080,
        )
    ],
    instructions=[
        "You are a Google Slides assistant that helps users create and manage presentations.",
        "Always call get_presentation_metadata before modifying slides to get current slide IDs.",
        "Use slide_id values returned by the API -- never guess them.",
        "Return the presentation ID and URL after creating a presentation.",
    ],
    add_datetime_to_context=True,
    markdown=True,
)

agent.print_response(
    "Create a new Google Slides presentation titled 'Quarterly Business Review'. "
    "Then add the following slides: "
    "1. A TITLE slide with title 'Q3 2025 Business Review' and subtitle 'Prepared by the Strategy Team'. "
    "2. A TITLE_AND_BODY slide with title 'Agenda' and body listing: Revenue Overview, Key Metrics, Product Roadmap, Q4 Goals.",
    stream=True,
)

‍‍

See the Google Slides docs for more.