Drive Document Reader

Reads and summarizes large documents from Google Drive.

document_reader.py
"""
Drive Document Reader
=====================
Reads and summarizes large documents from Google Drive.

Uses max_read_size to control the maximum file size loaded into memory
and returns structured summaries with key sections.

Key concepts:
- read_file: Exports Google Docs as text, Sheets as CSV, Slides as text
- add_datetime_to_context: Agent knows today's date for time-relative queries

Setup:
1. Create OAuth credentials at https://console.cloud.google.com (enable Google Drive API)
2. Export GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, GOOGLE_PROJECT_ID env vars
3. pip install openai google-api-python-client google-auth-httplib2 google-auth-oauthlib
4. First run opens browser for OAuth consent, saves token.json for reuse
"""

from agno.agent import Agent
from agno.models.openai import OpenAIResponses
from agno.tools.google.drive import GoogleDriveTools

# 50 MB — allow reading larger non-Workspace files (default is 10 MB)
MAX_READ_SIZE = 50 * 1024 * 1024

agent = Agent(
    name="Document Reader",
    model=OpenAIResponses(id="gpt-5.5"),
    tools=[GoogleDriveTools(max_read_size=MAX_READ_SIZE)],
    instructions=[
        "When reading documents, provide a structured summary with sections and key points.",
        "For spreadsheets (returned as CSV), describe the columns and highlight notable data.",
        "If the content is truncated, tell the user and summarize what was available.",
    ],
    add_datetime_to_context=True,
    markdown=True,
)

# ---------------------------------------------------------------------------
# Run Demo
# ---------------------------------------------------------------------------

if __name__ == "__main__":
    # Search and read a document
    agent.print_response(
        "Find the most recent Google Doc in my Drive and summarize it",
        stream=True,
    )

    # Read a specific file by ID
    # agent.print_response(
    #     "Read the file with ID <FILE_ID> and give me a detailed summary",
    #     stream=True,
    # )

    # Read a spreadsheet
    # agent.print_response(
    #     "Find a spreadsheet named 'Budget' and describe what data it contains",
    #     stream=True,
    # )

max_read_size is checked against non-Workspace file metadata before download. It neither expands Google Workspace's export limit nor controls the model's context window. Oversized files return an error rather than a truncated text result. Google Docs and Slides export as text, Sheets as CSV; the optional Office packages in setup extract DOCX, XLSX and PPTX. Binary formats such as PDF are not read as text by this tool.

Run the Example

Set up your virtual environment

uv venv --python 3.12
source .venv/bin/activate

Install dependencies

uv pip install -U agno google-api-python-client google-auth google-auth-httplib2 google-auth-oauthlib openai openpyxl python-docx python-pptx

Configure Google OAuth

Enable the Google API used by this example, configure the consent screen, and create a Desktop OAuth client in your Cloud project. Export its GOOGLE_CLIENT_ID, GOOGLE_CLIENT_SECRET, and GOOGLE_PROJECT_ID, or place the downloaded client JSON at credentials.json in the directory where you run Python. Google's Python quickstart shows the Desktop client setup.

The first tool call opens a browser for consent and caches credentials in token.json. Use a separate token_path when switching accounts. Passing an Agno user_id does not switch the authenticated Google account.

Export environment variables

export GOOGLE_CLIENT_ID="your_google_client_id_here"
export GOOGLE_CLIENT_SECRET="your_google_client_secret_here"
export GOOGLE_PROJECT_ID="your_google_project_id_here"
export OPENAI_API_KEY="your_openai_api_key_here"

Run the example

Save the code above as document_reader.py, then run:

python document_reader.py

Full source: cookbook/91_tools/google/drive/document_reader.py