
Google Cloud Spanner
Generally AvailableManage Cloud Spanner instances and databases and run multi-model (SQL + graph GQL) queries — governed by Cloud IAM.
- Database
- Distributed SQL
- Graph
- Authentication
- OAuth 2.0
- Hosting
- Vendor-hosted endpoint
- Required scopes
- 2
- Data actions
- Read and write
How it works
Every agent request passes through Ferentin policy before it reaches Google Cloud Spanner.
Ferentin's Google Cloud Spanner MCP server connects agents through OAuth2 + Cloud IAM, with tools for instance/config/database management, DDL schema evolution, and session-based query execution. Graph (GQL) and semantic queries run through the SQL-execution tools. SQL calls are auto request-tagged for auditing.
- AI agentAsks for something
- Ferentin policies
- OAuth 2.0 consent
- Cloud iam authorization
- No API keys
- Audit logging
- Google Cloud Spanner MCP serverAuthorized connection
- Governed resultMasked, logged, returned
What agents can do
15 tools · you choose which ones agents may callSearch and browse what exists, without opening the content itself.
- list_instances
- list_configs
- list_databases
Permissions requested
The OAuth 2.0 scopes Google Cloud Spanner asks you to approve when you connect.
| Scope | Access |
|---|---|
| https://www.googleapis.com/auth/spanner.admin | Admin |
| https://www.googleapis.com/auth/spanner.data | Not rated |
Before you connect
- Google does not support Dynamic Client Registration (DCR) — each tenant must create its own OAuth client in Google Cloud Console.
- Authentication is OAuth 2.0 + Google Cloud IAM. API keys are not accepted; access is enforced by the signing-in identity's IAM roles.
- Use access_type=offline to receive a refresh token for long-lived access.
- Requires roles/mcp.toolUser, roles/oauthconfig.editor, and Cloud Spanner Admin.
3 more technical notes
- A x-goog-user-project header attributes quota/billing to the project.
- Graph (GQL) and semantic/vector queries run through execute_sql / execute_sql_readonly — there are no separate graph tools.
- SQL executions are auto request-tagged (mcp_execute_sql, mcp_execute_sql_readonly, mcp_commit) for observability.
About Google Cloud Spanner
Cloud Spanner is Google Cloud's globally-distributed, strongly-consistent relational database with unified multi-model support (relational SQL, Spanner Graph GQL, and semantic/vector data).
Frequently asked about Google Cloud Spanner
- What can AI agents do with Google Cloud Spanner through Ferentin?
- Ferentin exposes 15 Google Cloud Spanner MCP tools to your AI agents, grouped as Discover (3 tools), Access (4 tools) and Manage (7 tools). Every call is checked against your Ferentin policies before it reaches Google Cloud Spanner.
- How is Google Cloud Spanner access secured?
- Requests through the Google Cloud Spanner integration are covered by OAuth 2.0 consent, cloud iam authorization, no API keys and audit logging. These controls are applied by Ferentin, not by Google Cloud Spanner, so they hold across every connected agent.
- What permissions does the Google Cloud Spanner integration require?
- The Google Cloud Spanner integration requests 2 OAuth 2.0 scopes: https://www.googleapis.com/auth/spanner.admin (admin access); https://www.googleapis.com/auth/spanner.data
- How do I connect Google Cloud Spanner to Ferentin?
- Open the Ferentin admin console, add Google Cloud Spanner as a new connection and authenticate with OAuth 2.0. You then choose which permissions and which tools your agents may use, and the connection goes live for everyone covered by your policies.
- Is the Google Cloud Spanner integration generally available?
- Yes. The Google Cloud Spanner integration is generally available on Ferentin, including the free tier.
Related integrations
3
Google Cloud SQL
Natural-language database operations on Cloud SQL (PostgreSQL/MySQL/SQL Server): run SQL, manage instances, users, and backups.

Google AlloyDB for PostgreSQL
Manage AlloyDB for PostgreSQL and run SQL, DDL, and vector-similarity search — governed by Cloud IAM.

Google BigQuery
Query and explore BigQuery data warehouses. Discover datasets/tables, inspect schema, and run read-only SQL — governed by Cloud IAM.