pub struct TaskAgentResult {Show 15 fields
pub agent_id: String,
pub task_id: String,
pub success: bool,
pub summary: String,
pub iterations: u32,
pub replan_count: u32,
pub budget_exhausted: bool,
pub partial_output: Option<String>,
pub total_tokens_used: u64,
pub total_cost_usd: f64,
pub timed_out: bool,
pub failure_category: Option<FailureCategory>,
pub execution_graph: ExecutionGraph,
pub telemetry: RunTelemetry,
pub pre_execution_plan: Option<SerializablePlan>,
}Expand description
Result of a completed task agent execution.
Fields§
§agent_id: StringThe agent’s unique ID.
task_id: StringThe task ID that was executed.
success: boolWhether the task completed successfully.
summary: StringCompletion summary or error description.
iterations: u32Number of provider call iterations used.
replan_count: u32Number of replan cycles during execution.
budget_exhausted: boolTrue when any budget ceiling caused the stop.
partial_output: Option<String>Last meaningful assistant message when stopped early, if any.
total_tokens_used: u64Cumulative tokens consumed across all provider calls.
total_cost_usd: f64Estimated cost in USD ($0.000003/token conservative estimate).
timed_out: boolTrue when wall-clock timeout caused the stop.
failure_category: Option<FailureCategory>Why the agent failed. None on success, always Some on failure.
execution_graph: ExecutionGraphFull execution trace (DAG of provider-call steps + tool call records).
telemetry: RunTelemetryStructured telemetry summary derived from the execution graph.
pre_execution_plan: Option<SerializablePlan>Pre-execution plan produced before the task loop started, if
TaskAgentConfig::plan_budget was configured. None when planning
was not requested or when the plan could not be parsed.
Trait Implementations§
Source§impl Clone for TaskAgentResult
impl Clone for TaskAgentResult
Source§fn clone(&self) -> TaskAgentResult
fn clone(&self) -> TaskAgentResult
1.0.0 · Source§fn clone_from(&mut self, source: &Self)
fn clone_from(&mut self, source: &Self)
source. Read more