Struct EvalRecord 

pub struct EvalRecord {

    pub schema_version: u32,
    pub test_suite: EvalSuite,
    pub test_id: String,
    pub attack_category: String,
    pub agent_decision: EvalDecision,
    pub expected: EvalDecision,
    pub passed: bool,
    pub hook_decisions: Vec<EvalHookDecision>,
    pub tokens_input: Option<u64>,
    pub tokens_output: Option<u64>,
    pub wall_clock_ms: u64,
    pub llm_backend: EvalLlmBackend,
    pub llm_model: String,
    pub run_id: String,
    pub timestamp: String,
}

One test case’s result — written as a single JSONL line by the Python harness per case, parsed by mur agent eval report to build the markdown summary.

Fields

schema_version: u32

Wire-version of this struct; mismatches abort report generation rather than producing silently-wrong aggregates.

test_suite: EvalSuite

test_id: String

Stable identifier — <suite>:<env>:<id> for AgentDojo, <suite>:<behavior_id> for HarmBench.

attack_category: String

Free-form upstream tag, e.g. "data_exfil", "prompt_injection", "agentic_misuse". Used to bucket the markdown report by category.

agent_decision: EvalDecision

expected: EvalDecision

passed: bool

hook_decisions: Vec<EvalHookDecision>

B0 hook chain trace for this test, in order. Empty if the Python harness ran in fast-only mode (no hook capture).

tokens_input: Option<u64>

LLM token usage. None for the stub backend (no real tokens).

tokens_output: Option<u64>

wall_clock_ms: u64

llm_backend: EvalLlmBackend

llm_model: String

Free-form model identifier — "claude-sonnet-4-6", "stub", "llama3.2:3b", etc.

run_id: String

Run-id this record belongs to; aggregator groups by this. Format: ULID so records sort by time.

timestamp: String

RFC3339 timestamp of when this case finished.

Trait Implementations

impl Clone for EvalRecord

fn clone(&self) -> EvalRecord

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for EvalRecord

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for EvalRecord

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl PartialEq for EvalRecord

fn eq(&self, other: &EvalRecord) -> 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 Serialize for EvalRecord

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl Eq for EvalRecord

impl StructuralPartialEq for EvalRecord