Skip to main content
Beta: This page contains some beta features and may change.
The Google Cloud action pack contains the following actions which enables users to query live data in Google BigQuery via SQL and get the results in a tabular format:
  • Search BigQuery with SQL: BigQuery read-only, user OAuth, SQL text input.
  • Gemini BigQuery Analytics with Context: Natural-language analytics, user OAuth (cloud-platform scope), centralized OAuth app with background execution supported by default.
  • Gemini Looker Analytics with Context: Natural-language analytics for Looker Explores, user OAuth, centralized OAuth app.
  • Gemini Looker Studio Analytics with Context: Natural-language analytics for Looker Studio datasources, user OAuth.
  • Gemini Analytics with Data Agent: Natural-language analytics via pre-configured conversational data agents.

​
Key features

  • Real-time warehouse access: No extra indexing or data duplication required.
  • Permission-aware search: User-level OAuth ensures results match the caller’s existing BigQuery permissions.
  • Read-only guarantees: All write and DDL statements are blocked by the bigquery.readonly OAuth scope.
  • Unified intelligence: Glean Agents combine BigQuery results with unstructured documents to generate contextual, unified answers.

​
Prerequisites

This action pack requires the following Google APIs to be enabled:

​
Setup Google Cloud action pack

This action pack needs the GCP Project ID that you want to bill your queries on. BigQuery determines the location to run the job based on the datasets referenced in the request.
  1. Open the Glean Admin Console → Platform → Actions.
  2. Click Add and select Google Cloud Actions.
  3. Enter the GCP Project ID in the Configuration section.
  4. In Publish Settings, choose the actions that you want to enable in Agents.
  5. Click Save.
Glean’s OAuth App is currently under review and should be verified soon. Until then admins have to authorize the unverified app. Otherwise users will see a message that the app is unverified when authenticating with this option.

​
Additional setup for Gemini Looker Analytics with Context action

You need to add the Looker Instance Qualifier and the Looker Client ID in the Looker Configuration section for the Action Pack. The Looker Instance field takes the qualifier for your instance (the part before looker.com). For example, if your instance has a URI like sampleinstance.cloud.looker.com, provide sampleinstance.cloud in the field. The Looker Client ID is the ID of the API Client that you have set up for Glean to access your instance. To set up a Client for Glean:
  1. Install the API Explorer in Looker.
  2. Register an OAuth Client Application with the API Explorer. The following parameters must be set:
    • client_guid: A globally unique ID for the application. For example, company_name-app-for-glean.
    • redirect_uri: Copy the redirect uri from the Action Pack configuration page.
    • display_name: The name of the application that is displayed to users of the application.
    • description: A description of the application that is displayed to users on a disclosure and confirmation page when a user first logs in from the application.
    • enabled: true
  3. Provide the client_guid in the Client ID field in Glean.

​
Additional setup for Gemini Analytics with Data Agent action

Data Agents need to be set up with appropriate context and system instructions using the Conversational Analytics API. Any user who wants to create a data agent must have the Gemini Data Analytics Data Agent Creator role.

​
Usage instructions

​
Set Up in Agent Builder

  1. Add a step and navigate to Select Step > All actions > By data source > Google Cloud.
  2. Select one of these actions:
    • To execute direct SQL to BigQuery:
      • Select Search BigQuery with SQL
      • Enter the SQL that you wish to execute. Example: select * from acmecorp.customers.dim_all_customers
    • To issue queries in natural language to BigQuery:
      • Select Gemini BigQuery Analytics with Context
      • Provide the fully-qualified table names to be used as context in a comma-separated list. Example: acmecorp.orders.daily_sales,acmecorp.customers.dim_all_customers
      • Enter a natural language instruction. Example: Which customer has the highest daily sales?
    • To issue queries in natural language to Looker:
      • Select Gemini Looker Analytics with Context
      • Provide the fully-qualified Looker Explores to be used as context. These can be obtained from the URL as you open an Explore. For example, if the URL is https://acmecorp.cloud.looker.com/explore/LookML1/Explore1 then the Looker ML model is LookML1.Explore1
      • Enter a natural language instruction. Example: Use the datasources to answer this query: [[query]]
    • To issue queries in natural language to Looker Studio:
      • Select Gemini Looker Studio Analytics with Context
      • Provide the datasource IDs to be used as context in a comma-separated list. For example, if the URL for a datasource is https://lookerstudio.google.com/datasources/c6b5b53c-c261-4a64-b136-0b96b5d9daab, then the datasource ID is: c6b5b53c-c261-4a64-b136-0b96b5d9daab
      • Enter a natural language instruction. Example: Analyze the following query with the datasources provided: [[query]]
    • To issue queries in natural language to a Data Agent:
      • Select Gemini Analytics with Data Agent
      • Provide the ID of the Data Agent. This is the last segment in the Data Agent name. For example, for projects/acmecorp/locations/global/dataAgents/data-agent-1, the ID is data-agent-1
      • Enter a natural language instruction. Example: Answer this query: [[query]]

​
Grant OAuth access

Because Glean ships with a central, scope-restricted Google OAuth client, customers do not have to create their own. Users simply:
  1. Run an agent that calls a Google Cloud action.
  2. Follow the Google consent screen and grant permission:

​
Best practices and troubleshooting

  • Always supply a valid Project ID; otherwise calls will fail with a 404 error.
  • If your queries time out, you can reach out to your Glean account team to raise it. We recommend keeping queries under 120 secs to avoid BigQuery cost spikes.
  • If a natural‑language question is ambiguous, include short hints like datasets, measures or filters to improve quality and reduce retries.
  • Specify the least-possible-sized ID list when using Gemini Analytics with inline context to ensure maximum quality.
  • Gemini Analytics currently require the broad cloud-platform scope.

​
Known limitations

  • Only SELECT statements are supported. Write statements (INSERT/UPDATE/DDL) will return “read-only scope” errors.
  • SQL results longer than 10 rows are truncated in the current UI. We advise adding a LIMIT or filter.
  • Multi-turn conversations are not preserved; each action run is stateless.
  • No automatic retries on BigQuery quota or transient errors – rerun the agent if that occurs.
  • See also: Gemini for Google Cloud - Conversational Analytics API known limitations