pub struct PreparedRun {
pub graph: Graph,
pub spec: StateSpec,
pub initial_state: Option<State>,
}Expand description
Everything the runner needs to execute one case repetition.
Built by the agent factory per run: graphs are cheap to compile, and a
fresh build per repetition guarantees no state leaks between runs. The
factory receives the run’s Journal so recording-wired graphs (e.g.
create_react_agent_with_recording) journal their model and tool calls
into the run’s evidence.
Fields§
§graph: GraphThe compiled graph under test.
spec: StateSpecThe state schema (channels + reducers).
initial_state: Option<State>Explicit initial state. None derives it from the case’s input
payload (which must then be a JSON object of channel values).
Implementations§
Source§impl PreparedRun
impl PreparedRun
Auto Trait Implementations§
impl !RefUnwindSafe for PreparedRun
impl !UnwindSafe for PreparedRun
impl Freeze for PreparedRun
impl Send for PreparedRun
impl Sync for PreparedRun
impl Unpin for PreparedRun
impl UnsafeUnpin for PreparedRun
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more