Struct AgentRunSpec 

pub struct AgentRunSpec {
    pub workflow_id: WorkflowId,
    pub activity_id: ActivityId,
    pub attempt: u32,
    pub activity_type: String,
    pub input: Payload,
}

The neutral identity + input for one agent run (one activity attempt).

Carries only what every harness needs to run an attempt: the (workflow, activity, attempt) key, the activity-type name the engine dispatched, and the input Payload. It names no harness-specific configuration — an adapter holds any harness-specific settings itself (constructed when the crate::AgentHarness is built), keeping this spec harness-blind.

Fields

workflow_id: WorkflowId

The workflow this activity attempt belongs to.

activity_id: ActivityId

The activity within the workflow.

attempt: u32

The attempt number of the activity being run.

activity_type: String

The activity-type name the engine dispatched to this run (e.g. "dev").

Neutral run identity, not harness configuration: it is the same type name the workflow definition routed on, so an adapter can label or parameterise a run by what the engine asked for without the spec learning anything harness-specific.

input: Payload

The activity input handed to the agent.

Implementations

impl AgentRunSpec

pub fn new( workflow_id: WorkflowId, activity_id: ActivityId, attempt: u32, activity_type: impl Into<String>, input: Payload, ) -> Self

Builds a run spec from the neutral run identity and input payload.

Trait Implementations

impl Clone for AgentRunSpec

fn clone(&self) -> AgentRunSpec

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AgentRunSpec

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

Formats the value using the given formatter.

impl Eq for AgentRunSpec

impl PartialEq for AgentRunSpec

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

Equality operator ==.

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

Inequality operator !=.

impl StructuralPartialEq for AgentRunSpec