Struct TaskHandle 

pub struct TaskHandle { /* private fields */ }

Handle for driving a tracked task to a terminal state.

Implementations

impl TaskHandle

pub const fn id(&self) -> &TaskId

Get the task ID.

pub fn task(&self) -> Option<Task>

A snapshot of this task’s current state.

pub fn cancel_token(&self) -> Option<CancellationToken>

The cancellation token for this task, for wiring into an execution context so tasks/cancel aborts the running operation.

pub fn mark_input_required(&self) -> Result<(), McpError>

Mark the task as waiting for input (e.g. during elicitation/sampling).

pub fn complete(&self, payload: Value) -> Result<(), McpError>

Mark the task completed and store its payload.

pub fn fail(&self, message: impl Into<String>) -> Result<(), McpError>

Mark the task failed with a status message.

The stored tasks/result payload is an internal JSON-RPC error carrying message. When the underlying request failed with a specific JSON-RPC error, prefer fail_with_error so tasks/result reproduces it exactly.

pub fn fail_with_error(&self, error: JsonRpcError) -> Result<(), McpError>

Mark the task failed, storing the JSON-RPC error the underlying request would have returned (reproduced verbatim by tasks/result).

pub fn fail_with_result( &self, payload: Value, message: Option<String>, ) -> Result<(), McpError>

Mark the task failed while storing a successful payload.

Per spec, a tools/call whose result has isError: true reaches the failed status, but tasks/result still returns that result.

pub fn is_cancelled(&self) -> bool

Check if the task has been cancelled.

pub async fn cancelled(&self)

A future that completes when the task is cancelled.