How do AI agents use Google Cloud Search through Open Connector?

Your user connects Google Cloud Search once with one of its cataloged authentication methods. Open Connector stores the credential in an encrypted vault and exposes Google Cloud Search tools to your agent over MCP or a typed API, with credentials injected server-side on each call.

Is this a Google Cloud Search MCP server?

Yes. Open Connector can serve Google Cloud Search as a named MCP server with a scoped allowlist and a per-user connection URL, so any MCP client can call Google Cloud Search actions with credentials injected server-side.

Where do Google Cloud Search credentials live?

In your own infrastructure. Open Connector keeps credentials in its own vault and injects them at call time, so they never leave your environment.

Google Cloud Search integration

Google Cloud Search integration for AI agents.

Google Cloud Search integration for AI agents with secure authentication and server-side credential injection. Open Connector runs the OAuth, seals the token in an encrypted vault, and serves Google Cloud Search tools to your agent over MCP or a typed API — credentials injected server-side, every call audited, nothing leaving your infrastructure. Open source (AGPL-3.0) and self-hostable.

View source on GitHub All integrations →

What your agents can do

Real Google Cloud Search actions, managed and audited.

Your user connects Google Cloud Search once; your agent can then secure AI-agent actions — scoped to the OAuth permissions you grant and the tool allowlist you configure. Every action is least-privilege and written to a tamper-evident audit trail.

1
Your user grants Google Cloud Search access once (OAuth) — the token lands in the vault.
2
Your agent calls a tool over MCP or the typed API; Open Connector injects the credential server-side.
3
Every routed call appends a hash-chained audit record — nothing leaves your infra.

Tools & triggers

Supported Google Cloud Search tools.

44 tools are generated from the published Google Cloud Search catalog. Descriptions are plain text; each action remains subject to its configured authentication and tool allowlist.

Showing 44 tools. All published catalog entries are included in this page's server-rendered HTML.

Cloudsearch Debug Datasources Items CheckAccess: Checks whether an item is accessible by specified principal. Principal must be a user; groups and domain values aren't supported. **Note:** This API requires an admin account to execute.; debug
Cloudsearch Debug Datasources Items SearchByViewUrl: Fetches the item whose viewUrl exactly matches that of the URL provided in the request. **Note:** This API requires an admin account to execute.; debug
Cloudsearch Debug Datasources Items Unmappedids List: List all unmapped identities for a specific item. **Note:** This API requires an admin account to execute.; debug
Cloudsearch Debug Identitysources Items ListForunmappedidentity: Lists names of items associated with an unmapped identity. **Note:** This API requires an admin account to execute.; debug
Cloudsearch Indexing Datasources DeleteSchema: Deletes the schema of a data source. **Note:** This API requires an admin or service account to execute.; indexing
Cloudsearch Indexing Datasources GetSchema: Gets the schema of a data source. **Note:** This API requires an admin or service account to execute.; indexing
Cloudsearch Indexing Datasources Items Delete: Deletes Item resource for the specified resource name. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items DeleteQueueItems: Deletes all items in a queue. This method is useful for deleting stale items. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Get: Gets Item resource by item name. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Index: Updates Item ACL, metadata, and content. It will insert the Item if it does not exist. This method does not support partial updates. Fields with no provided values are cleared out in the Cloud Search index. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items List: Lists all or a subset of Item resources. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Poll: Polls for unreserved items from the indexing queue and marks a set as reserved, starting with items that have the oldest timestamp from the highest priority ItemStatus. The priority order is as follows: ERROR MODIFIED NEW_ITEM ACCEPTED Reserving items ensures that polling from other threads cannot create overlapping sets. After handling the reserved items, the client should put items back into the unreserved state, either by calling index, or by calling push with the type REQUEUE. Items automatically become available (unreserved) after 4 hours even if no update or push method is called. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Push: Pushes an item onto a queue for later polling and updating. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Unreserve: Unreserves all items from a queue, making them all eligible to be polled. This method is useful for resetting the indexing queue after a connector has been restarted. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources Items Upload: Creates an upload session for uploading item content. For items smaller than 100 KB, it's easier to embed the content inline within an index request. This API requires an admin or service account to execute. The service account used is the one whitelisted in the corresponding data source.; indexing
Cloudsearch Indexing Datasources UpdateSchema: Updates the schema of a data source. This method does not perform incremental updates to the schema. Instead, this method updates the schema by overwriting the entire schema. **Note:** This API requires an admin or service account to execute.; indexing
Cloudsearch InitializeCustomer: Enables `third party` support in Google Cloud Search. **Note:** This API requires an admin account to execute.; cloudsearch
Cloudsearch Media Upload: Uploads media for indexing. The upload endpoint supports direct and resumable upload protocols and is intended for large items that can not be [inlined during index requests](https://developers.google.com/workspace/cloud-search/docs/reference/rest/v1/indexing.datasources.items#itemcontent). To index large content: 1. Call indexing.datasources.items.upload with the item name to begin an upload session and retrieve the UploadItemRef. 1. Call media.upload to upload the content, as a streaming request, using the same resource name from the UploadItemRef from step 1. 1. Call indexing.datasources.items.index to index the item. Populate the [ItemContent](/cloud-search/docs/reference/rest/v1/indexing.datasources.items#ItemContent) with the UploadItemRef from step 1. For additional information, see [Create a content connector using the REST API](https://developers.google.com/workspace/cloud-search/docs/guides/content-connector#rest). **Note:** This API requires a service account to execute.; media
Cloudsearch Operations Get: Gets the latest state of a long-running operation. Clients can use this method to poll the operation result at intervals as recommended by the API service.; operations
Cloudsearch Operations Lro List: Lists operations that match the specified filter in the request. If the server doesn't support this method, it returns `UNIMPLEMENTED`.; operations
Cloudsearch Query RemoveActivity: Provides functionality to remove logged activity for a user. Currently to be used only for Chat 1p clients **Note:** This API requires a standard end user account to execute. A service account can't perform Remove Activity requests directly; to use a service account to perform queries, set up [Google Workspace domain-wide delegation of authority](https://developers.google.com/workspace/cloud-search/docs/guides/delegation/).; query
Cloudsearch Query Search: The Cloud Search Query API provides the search method, which returns the most relevant results from a user query. The results can come from Google Workspace apps, such as Gmail or Google Drive, or they can come from data that you have indexed from a third party. **Note:** This API requires a standard end user account to execute. A service account can't perform Query API requests directly; to use a service account to perform queries, set up [Google Workspace domain-wide delegation of authority](https://developers.google.com/workspace/cloud-search/docs/guides/delegation/).; query
Cloudsearch Query Sources List: Returns list of sources that user can use for Search and Suggest APIs. **Note:** This API requires a standard end user account to execute. A service account can't perform Query API requests directly; to use a service account to perform queries, set up [Google Workspace domain-wide delegation of authority](https://developers.google.com/workspace/cloud-search/docs/guides/delegation/).; query
Cloudsearch Query Suggest: Provides suggestions for autocompleting the query. **Note:** This API requires a standard end user account to execute. A service account can't perform Query API requests directly; to use a service account to perform queries, set up [Google Workspace domain-wide delegation of authority](https://developers.google.com/workspace/cloud-search/docs/guides/delegation/).; query
Cloudsearch Settings Datasources Create: Creates a datasource. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Datasources Delete: Deletes a datasource. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Datasources Get: Gets a datasource. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Datasources List: Lists datasources. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Datasources Patch: Updates a datasource. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Datasources Update: Updates a datasource. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings GetCustomer: Get customer settings. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Searchapplications Create: Creates a search application. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Searchapplications List: Lists all search applications. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings Searchapplications Reset: Resets a search application to default settings. This will return an empty response. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Settings UpdateCustomer: Update customer settings. **Note:** This API requires an admin account to execute.; settings
Cloudsearch Stats GetIndex: Gets indexed item statistics aggreggated across all data sources. This API only returns statistics for previous dates; it doesn't return statistics for the current day. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats GetQuery: Get the query statistics for customer. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats GetSearchapplication: Get search application stats for customer. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats GetSession: Get the # of search sessions, % of successful sessions with a click query statistics for customer. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats GetUser: Get the users statistics for customer. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats Index Datasources Get: Gets indexed item statistics for a single data source. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats Query Searchapplications Get: Get the query statistics for search application. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats Session Searchapplications Get: Get the # of search sessions, % of successful sessions with a click query statistics for search application. **Note:** This API requires a standard end user account to execute.; stats
Cloudsearch Stats User Searchapplications Get: Get the users statistics for search application. **Note:** This API requires a standard end user account to execute.; stats

FAQ

Google Cloud Search integration, answered

How do AI agents use Google Cloud Search through Open Connector?: Your user connects Google Cloud Search once with one of its cataloged authentication methods. Open Connector stores the credential in an encrypted vault and exposes Google Cloud Search tools to your agent over MCP or a typed API, with credentials injected server-side on each call.
Is this a Google Cloud Search MCP server?: Yes. Open Connector can serve Google Cloud Search as a named MCP server with a scoped allowlist and a per-user connection URL, so any MCP client can call Google Cloud Search actions with credentials injected server-side.
Where do Google Cloud Search credentials live?: In your own infrastructure. Open Connector keeps credentials in its own vault and injects them at call time, so they never leave your environment.

Give your agents Google Cloud Search — keep the keys.

Open source, self-hostable, with Google Cloud Search credentials that never leave your infrastructure. Run it from source today.

View source on GitHub How the MCP gateway works →