Struct TraceScore 

pub struct TraceScore {

    pub public_id: TraceScoreId,
    pub internal_id: Uuid,
    pub org_id: i64,
    pub observer_id: ObserverId,
    pub scorer_key: String,
    pub session_id: SessionId,
    pub turn_id: String,
    pub agent_id: Option<AgentId>,
    pub agent_version_id: Option<AgentVersionId>,
    pub harness_id: Option<HarnessId>,
    pub status: TraceScoreStatus,
    pub pass: Option<bool>,
    pub value: Option<f64>,
    pub label: Option<String>,
    pub reason: Option<String>,
    pub judge_input_tokens: Option<u64>,
    pub judge_output_tokens: Option<u64>,
    pub judge_cost_usd: Option<f64>,
    pub error_message: Option<String>,
    pub created_at: DateTime<Utc>,
    pub updated_at: DateTime<Utc>,
}

One score produced by an observer scorer for one trace slice. Linked back to the exact session/turn it graded; agent/harness identifiers are denormalized at scoring time for aggregation.

Fields

public_id: TraceScoreId

External identifier (score_<32-hex>). Shown as “id” in API.

internal_id: Uuid

Internal UUID primary key. Never exposed in API.

org_id: i64

Organization ID. Internal only.

observer_id: ObserverId

Observer that produced this score.

scorer_key: String

Scorer key within the observer.

session_id: SessionId

Session this score grades.

turn_id: String

Turn this score grades (turn scope).

agent_id: Option<AgentId>

Agent active in the session at scoring time.

agent_version_id: Option<AgentVersionId>

Agent version active in the session at scoring time.

harness_id: Option<HarnessId>

Harness of the session.

status: TraceScoreStatus

pass: Option<bool>

Whether the scorer passed (set when completed).

value: Option<f64>

Score value 0.0–1.0 (set when completed).

label: Option<String>

Optional categorical label from an LLM judge (e.g. missing_source).

reason: Option<String>

Human-readable explanation (set when completed). For LLM judges this is the judge’s reasoning — retained as the raw material for the Phase 2 improvement loop.

judge_input_tokens: Option<u64>

Judge LLM input tokens (llm_judge scores only).

judge_output_tokens: Option<u64>

Judge LLM output tokens (llm_judge scores only).

judge_cost_usd: Option<f64>

Judge call cost in USD when the provider reports it (llm_judge only).

error_message: Option<String>

Error details if errored.

created_at: DateTime<Utc>

updated_at: DateTime<Utc>

Trait Implementations

impl Clone for TraceScore

fn clone(&self) -> TraceScore

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for TraceScore

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for TraceScore

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for TraceScore

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.