BigQuery function call succeeds locally but fails silently on Vertex AI

[KSonPham]

Hi,

I’m building an agent that queries a BigQuery table using the following function:

def run_bigquery_sql(sql_query: str) -> str:
    client = bigquery.Client()
    query_job = client.query(sql_query)
    results = query_job.result()
    return json.dumps(results, indent=2, default=str)

This works fine locally. However, when deployed to Vertex AI, the function appears to run (i.e., no errors), but no data is returned. The model’s response confirms the function is invoked, but there’s no result.

---

Example call

User prompt:

Show me the type of partners and their count?

Model function call:

{
  "function_call": {
    "name": "run_bigquery_sql",
    "args": {
      "sql_query": "SELECT partner_type__c, COUNT(*) FROM `modified-talon-451917-8.ff_df_sf_partner.Partner_c_mini` GROUP BY partner_type__c"
    }
  }
}

---

Suspected Cause

Likely an authentication issue with the BigQuery client when running in Vertex AI (possibly missing service account permissions or not properly configured auth context).

---

Question

What is the correct way to authenticate a BigQuery client in a Vertex AI function/tool call so that it returns data as expected?

[shobsi]

Could you post the output of the function in the Vertex AI? I am not sure about the "works fine locally" part. The variable results is of type google.cloud.bigquery.table.RowIterator which you need to iterate to get the result.

PS: ADK is now shipping built-in BigQuery tools (code, sample agent)

[KSonPham]

Hi @shobsi thank you for answering. By works fine locally i mean the agent can use function call to get info from BigQuery and answer accordingly. This is the result of the VertexAI, as you can see only function call and SQL was generated:

{'content': {'parts': [{'function_call': {'id': 'adk-a92d4168-7ad6-4f5a-bf6f-75f9e3490945', 'args': {'sql_query': 'SELECT partner type__c, COUNT(*) FROM `modified-talon-451917-b8.ff_df_sf_partner.Partner_c_mini` GROUP BY Partnertyp__c'}, 'name': 'run_bigquery_sql'}}], 'role': 'model'}, 'usage_metadata': {'candidates_token_count': 54, 'candidates_tokens_details': [{'modality': 'TEXT', 'token_count': 54}], 'prompt_token_count': 1890, 'prompt_tokens_details': [{'modality': 'TEXT', 'token_count': 1890}], 'thoughts_token_count': 149, 'total_token_count': 2093, 'traffic_type': 'ON_DEMAND'}, 'invocation_id': 'e-93dfcb8f-7705-45f5-9635-935d0ce17b99', 'author': 'Data_agent', 'actions': {'state_delta': {}, 'artifact_delta': {}, 'requested_auth_configs': {}}, 'long_running_tool_ids': [], 'id': 'o2GeHG2D', 'timestamp': 1750240571.029617}
Query 3 response time: 2,789 seconds

Running remote query 4: "Show me top 10 most valuable partners using this year and last year revenue?"
{'content': {'parts': [{'function_call': {'id': 'adk-d9c31382-6285-49a0-9f9b-a66565bbabc9', 'args': {'sql_query': 'SELECT Company__c, (Sales_current_year__c + Sales_previous_year__c) AS Total_Revenue FROM `modified-talon-451917-b8.ff_df_sf_partner.Partner_c_mini` ORDER BY Total_Revenue DESC LIMIT 10'}, 'name': 'run_bigquery_sql'}}], 'role': 'model'}, 'usage_metadata': {'candidates_token_count': 78, 'candidates_tokens_details': [{'modality': 'TEXT', 'token_count': 78}], 'prompt_token_count': 1961, 'prompt_tokens_details': [{'modality': 'TEXT', 'token_count': 1961}], 'thoughts_token_count': 141, 'total_token_count': 2180, 'traffic_type': 'ON_DEMAND'}, 'invocation_id': 'e-9285b287-b98b-4860-841d-8347067f29e3', 'author': 'Data_agent', 'actions': {'state_delta': {}, 'artifact_delta': {}, 'requested_auth_configs': {}}, 'long_running_tool_ids': [], 'id': 'uWgSF8aN', 'timestamp': 1750240573.74284}
Query 4 response time: 2,510 seconds```