Enum ModelResponse 

pub enum ModelResponse {
    Message {
        text: String,
        stop_reason: String,
        usage: Option<HarnessUsage>,
    },
    ToolCall {
        preface: Option<String>,
        invocation: ToolInvocation,
        usage: Option<HarnessUsage>,
    },
}

Response from a single model call. Message is a final answer (no follow-up tool needed); ToolCall hands control back to the harness loop to execute the tool and feed the result back next turn.

usage carries the provider-reported token tally for this call. AgentLoopHarness accumulates these across all steps of a turn and attaches the total to the final HarnessInternalEvent::TurnEnd. None ⇒ provider didn’t report usage on this call (e.g. mid-stream chunk, scripted fake client).

Variants

Message

Fields

text: String

stop_reason: String

usage: Option<HarnessUsage>

ToolCall

Fields

preface: Option<String>

invocation: ToolInvocation

usage: Option<HarnessUsage>

Implementations

impl ModelResponse

pub fn usage(&self) -> Option<&HarnessUsage>

Per-call usage, regardless of variant. Pulled out so AgentLoopHarness can fold it into the running total without matching on every branch.

Trait Implementations

impl Clone for ModelResponse

fn clone(&self) -> ModelResponse

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ModelResponse

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

Formats the value using the given formatter.

impl PartialEq for ModelResponse

fn eq(&self, other: &ModelResponse) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for ModelResponse