Readonly[A unique symbol to identify ADK base tool class.
Readonly[Brand identifying this object as a NodeTool (see isNodeTool).
ReadonlydescriptionReadonlyisReadonlynameReadonlynodeThe Google API LLM variant to use.
Gets the OpenAPI specification of this tool in the form of a FunctionDeclaration.
NOTE
processLlmRequest to add function declaration to LLM request.The FunctionDeclaration of this tool, or undefined if it doesn't need to be added to LlmRequest.config.
Whether this tool needs a human to approve args before it runs.
The gate itself lives in the tool that owns it (see FunctionTool), but
the resume path has to ask the same question a turn later, to check that an
approval it is about to honour belongs to a tool that gates at all. A tool
that never gates returns false here, which is the safe default: an approval
naming it is meaningless and gets rejected rather than executed. Mirrors
Python's BaseTool.check_require_confirmation.
The arguments the tool would run with.
Optional_toolContext: ContextThe context of the call, when there is one.
Whether the call requires confirmation.
Processes the outgoing LLM request for this tool.
Use cases:
The request to process the LLM request.
Runs the tool with the given arguments and context.
NOTE
The request to run the tool.
A promise that resolves to the tool response.
A tool that executes a BaseNode (e.g. a
Workflowor a function node) on behalf of anLlmAgent. This is the inverse of ToolNode (which exposes a tool as a workflow node): here a node/workflow is exposed to a model as a callable tool.The wrapped node MUST declare an
inputSchema(the tool's parameter schema is derived from it). When the model calls the tool, the node runs with a NodeContext bridged from the tool's agent context (sharing the invocation, session, and state); the node's structured output becomes the tool result.Ported from
google/adk-pythontools/_node_tool.py::NodeTool.The tool is marked long-running so a node that pauses for input (
RequestInput) does not force a synthetic empty response.