	In Browser
	StumbleUpon
	del.icio.us
	Google
	Google Buzz
	reddit
	LinkedIn

	Facebook
	Twitter
	Linkedin
	E-Mail

Generative AI > Google Gemini API > ADK with Google Sheets - Reading and Writing Data

ADK with Google Sheets - Reading and Writing Data

Author: Venkata Sudhakar

Google Sheets is widely used in Indian businesses for inventory tracking, sales reporting, and budget management. An ADK agent connected to Sheets can read live data, perform analysis, and write results back - turning a static spreadsheet into a conversational reporting tool. ShopMax India uses this to let managers query the daily sales sheet in plain English and append summary rows automatically.

The Sheets API uses the spreadsheets scope. The agent tools use the spreadsheets.values.get method to read ranges and spreadsheets.values.append to write new rows. Service account credentials with editor access to the target spreadsheet are sufficient - no domain-wide delegation needed if the sheet is shared with the service account email.

The below example shows an ADK agent that reads sales data from a sheet and appends a daily summary row.

from google.adk.agents import LlmAgent
from google.adk.tools import FunctionTool
from google.adk.sessions import InMemorySessionService
from google.adk.runners import Runner
from google.oauth2 import service_account
from googleapiclient.discovery import build

SCOPES = ["https://www.googleapis.com/auth/spreadsheets"]
SHEET_ID = "1BxiMVs0XRA5nFMdKvBdBZjgmUUqptlbs74OgVE2upms"  # example ID

def get_sheets_service():
    creds = service_account.Credentials.from_service_account_file(
        "/secrets/sa-key.json", scopes=SCOPES
    )
    return build("sheets", "v4", credentials=creds).spreadsheets()

def read_sales_data(sheet_range: str = "Sales!A1:E20") -> dict:
    """Read sales data from the ShopMax daily sales sheet."""
    svc = get_sheets_service()
    result = svc.values().get(spreadsheetId=SHEET_ID, range=sheet_range).execute()
    rows = result.get("values", [])
    if not rows:
        return {"rows": []}
    headers = rows[0]
    data = [dict(zip(headers, row)) for row in rows[1:]]
    return {"row_count": len(data), "data": data}

def append_summary_row(sheet_name: str, values: list) -> dict:
    """Append a summary row to the specified sheet."""
    svc = get_sheets_service()
    body = {"values": [values]}
    result = svc.values().append(
        spreadsheetId=SHEET_ID,
        range=f"{sheet_name}!A1",
        valueInputOption="USER_ENTERED",
        insertDataOption="INSERT_ROWS",
        body=body,
    ).execute()
    return {"status": "appended", "updated_range": result.get("updates", {}).get("updatedRange")}

agent = LlmAgent(
    name="shopmax_sheets_agent",
    model="gemini-2.0-flash",
    instruction=(
        "You are a ShopMax India reporting agent. "
        "Read the sales sheet, compute the total revenue and top product, "
        "then append a one-line summary row with date, total, and top product."
    ),
    tools=[FunctionTool(read_sales_data), FunctionTool(append_summary_row)],
)

session_service = InMemorySessionService()
runner = Runner(agent=agent, session_service=session_service)
events = list(runner.run(
    user_id="reporting_bot", session_id="sess_sheets_001",
    new_message={"role": "user", "parts": [{"text": "Summarise today sales and append the summary to the sheet."}]},
))
print(next(e.content.parts[0].text for e in events if e.is_final_response()))

It gives the following output,

Read 18 sales rows from Sales!A1:E20.

Summary for 2026-04-06:
  Total Revenue : Rs 4,12,850
  Top Product   : Samsung 55" QLED TV  (Rs 98,400 across 6 units)
  Store Count   : 4 stores reported

Appended summary row to Sales sheet:
  [2026-04-06, SUMMARY, Rs 4,12,850, Samsung 55" QLED TV, 18 transactions]
  Updated range: Sales!A20

For automated daily reporting, deploy the agent as a Cloud Run job triggered by Cloud Scheduler at 11 PM IST. Use the Sheets API batchUpdate method to apply formatting (bold, colour) to summary rows. Chain this agent with the Gmail agent to email the summary to the regional manager every morning at 9 AM.

Send your comments, suggestions or queries regarding this site to [email protected].