Struct ToolResult 

pub struct ToolResult {
    pub content: String,
    pub is_error: bool,
    pub images: Vec<ToolImage>,
    pub is_pending: bool,
    pub pending_task_id: Option<String>,
}

Result returned to the LLM after tool execution.

Fields

content: String

Text content - LLM will read this.

is_error: bool

Whether this is an error (LLM can see errors and adjust strategy).

images: Vec<ToolImage>

Images attached to this result (e.g. from read tool reading an image file). Most tools leave this empty.

is_pending: bool

true when this is a placeholder for an async task: content tells the LLM the work is in progress, and the real result is reinjected later (by the Agent) when the background task finishes. The Agent uses this flag to emit AsyncToolStarted instead of treating the call as finished. The placeholder content is still added to history as the tool message so the LLM can continue other work while waiting.

pending_task_id: Option<String>

Task id of the background task, set iff is_pending. Used by the Agent to track/cancel the task and by the frontend to correlate progress.

Implementations

impl ToolResult

pub fn success(content: impl Into<String>) -> Self

pub fn error(content: impl Into<String>) -> Self

pub fn pending(content: impl Into<String>, task_id: String) -> Self

Build a pending placeholder for an async task. content should tell the LLM what is happening and the task_id it can reference later.

Trait Implementations

impl Clone for ToolResult

fn clone(&self) -> ToolResult

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for ToolResult

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.