Struct GoogleSearchTool 

pub struct GoogleSearchTool;

Available on crate feature tools only.

GoogleSearch is a built-in tool that is automatically invoked by Gemini models to retrieve search results from Google Search. The tool operates internally within the model and does not require or perform local code execution.

Implementations

impl GoogleSearchTool

pub fn new() -> GoogleSearchTool

Create a new GoogleSearchTool.

Trait Implementations

impl BypassMultiToolsLimit for GoogleSearchTool

Bypass support: convert the built-in Google Search tool into a function-calling tool so it can be used alongside custom function tools.

The Gemini Interactions API forbids mixing built-in (server-side) tools with custom function tools in one request. Implementing BypassMultiToolsLimit mirrors ADK-Python’s bypass_multi_tools_limit=True: the converted tool reports is_builtin() == false, declares a normal query: string function schema, and performs grounded search by delegating to an internal single-turn agent.

Because adk-tool cannot depend on adk-agent, the internal grounded-search agent is supplied by the caller. It is expected to be an LlmAgent configured with GoogleSearchTool and a Gemini model so that the grounding happens server-side.

Example

use adk_tool::{BypassMultiToolsLimit, GoogleSearchTool};
use std::sync::Arc;

// `search_agent` is an LlmAgent with GoogleSearchTool + a Gemini model.
let tool = GoogleSearchTool::new().with_bypass_multi_tools_limit(Arc::new(search_agent));
assert!(!tool.is_builtin());

fn bypass_name(&self) -> String

The function-tool name surfaced to the model after bypass conversion.

fn bypass_description(&self) -> String

The function-tool description surfaced to the model.

fn bypass_parameters_schema(&self) -> Value

The JSON Schema for the bypass function’s parameters.

fn bypass_query_field(&self) -> String

The argument field that carries the natural-language query forwarded to the internal agent (e.g. "query" for search, "url" for URL context).

fn with_bypass_multi_tools_limit(self, agent: Arc<dyn Agent>) -> Arc<dyn Tool>

Convert this built-in tool into a function-calling BypassBuiltinTool so it can coexist with custom function tools under the Interactions API.

impl Default for GoogleSearchTool

fn default() -> GoogleSearchTool

Returns the “default value” for a type.

impl Tool for GoogleSearchTool

fn name(&self) -> &str

Returns the unique name of this tool.

fn description(&self) -> &str

Returns a human-readable description of what this tool does.

fn is_builtin(&self) -> bool

Indicates whether this tool is a built-in server-side tool (e.g., google_search, url_context).

fn declaration(&self) -> Value

Returns the tool declaration that should be exposed to model providers.

fn execute<'life0, 'async_trait>( &'life0 self, _ctx: Arc<dyn ToolContext>, _args: Value, ) -> Pin<Box<dyn Future<Output = Result<Value, AdkError>> + Send + 'async_trait>>

where 'life0: 'async_trait, GoogleSearchTool: 'async_trait,

Executes the tool with the given context and arguments.

fn enhanced_description(&self) -> String

Returns an enhanced description that may include additional notes. For long-running tools, this includes a warning not to call the tool again if it has already returned a pending status. Default implementation returns the base description.

fn is_long_running(&self) -> bool

Indicates whether the tool is a long-running operation. Long-running tools typically return a task ID immediately and complete the operation asynchronously.

fn parameters_schema(&self) -> Option<Value>

Returns the JSON Schema for this tool’s parameters, if any.

fn response_schema(&self) -> Option<Value>

Returns the JSON Schema for this tool’s response, if any.

fn required_scopes(&self) -> &[&str]

Returns the scopes required to execute this tool.

fn is_read_only(&self) -> bool

Indicates whether this tool performs no side effects.

fn is_concurrency_safe(&self) -> bool

Indicates whether this tool is safe for concurrent execution.