google-adk-kotlin-core/com.google.adk.kt.tools/VertexAiSearchTool

VertexAiSearchTool

class VertexAiSearchTool(val dataStoreId: String? = null, val dataStoreSpecs: List<VertexAISearchDataStoreSpec>? = null, val searchEngineId: String? = null, val filter: String? = null, val maxResults: Int? = null, val bypassMultiToolsLimit: Boolean = false, val model: String? = null) : BaseTool

A built-in tool using Vertex AI Search.

This tool can be configured with either a dataStoreId (the Vertex AI search data store resource ID) or a searchEngineId (the Vertex AI search engine resource ID).

Constructors

VertexAiSearchTool

constructor(dataStoreId: String? = null, dataStoreSpecs: List<VertexAISearchDataStoreSpec>? = null, searchEngineId: String? = null, filter: String? = null, maxResults: Int? = null, bypassMultiToolsLimit: Boolean = false, model: String? = null)

Properties

bypassMultiToolsLimit

val bypassMultiToolsLimit: Boolean = false

When true, allows this built-in tool to be used alongside other tools: it is exposed as a function tool (see VertexAiSearchAgentTool) since built-in tools cannot otherwise be combined with other tools in a single request.

customMetadata

val customMetadata: Map<String, Any>

The custom metadata of the tool.

dataStoreId

val dataStoreId: String? = null

The Vertex AI search data store resource ID in the format of projects/{project}/locations/{location}/collections/{collection}/dataStores/{dataStore}.

dataStoreSpecs

val dataStoreSpecs: List<VertexAISearchDataStoreSpec>? = null

Specifications that define the specific DataStores to be searched. It should only be set if engine is used.

description

val description: String

The description of the tool.

filter

val filter: String? = null

The filter to apply to the search results.

isLongRunning

val isLongRunning: Boolean = false

Whether the tool's final result will be delivered out-of-band. When true, the framework marks the call as long-running and uses the tool's return value as the function-response payload. Returning Unit means "no response yet": the FR event is suppressed so the function-call event (which carries the call id in longRunningToolIds and is thus the turn's final response) ends the turn without re-invoking the model. A non-Unit return -- including an explicit empty Map -- is treated as a real response and emitted. (Unit suppression aligns with Python; Java instead always emits {}.) The longRunningToolIds id also drives the resumable-mode pause gate so the invocation can be resumed later via a user-injected function-response.