Struct Report 

pub struct Report {

    pub schema_version: u32,
    pub tool_version: String,
    pub lang: Lang,
    pub started_at: String,
    pub finished_at: String,
    pub duration_ms: u64,
    pub host: String,
    pub base_url: String,
    pub protocol: Protocol,
    pub model: String,
    pub claimed_model: String,
    pub depth: String,
    pub seed: u64,
    pub steps: Vec<String>,
    pub request_count: u32,
    pub results: Vec<ProbeResult>,
    pub verdict: Verdict,
    pub identity: Identity,
    pub billing: BillingAudit,
    pub channel: ChannelSignature,
    pub perf: PerfSummary,
    pub skipped: Vec<String>,
}

Fields

schema_version: u32

Shape of this JSON, independent of Report::tool_version.

A caller that stores reports needs to know whether an old row can still be deserialised, and the tool version cannot answer that: most releases change no field at all. Bumped only when a field is removed or its meaning changes; additive fields do not move it.

tool_version: String

lang: Lang

Language every human-readable string in this report is written in.

started_at: String

finished_at: String

duration_ms: u64

host: String

base_url: String

protocol: Protocol

model: String

claimed_model: String

depth: String

seed: u64

The random seed this run’s payloads were generated from. Replaying the same seed, depth and step list reproduces the same probes — which is what makes a verdict auditable after the fact.

Serialised as a string, and that is not cosmetic. A u64 seed routinely exceeds 2^53, and JSON numbers above that lose precision in every JavaScript consumer — a report rendered in a browser showed 2543944364647182727 as 25439443646471830000, next to a sentence promising the run could be replayed from it. A seed that cannot be copied accurately is worse than no seed, because it looks like evidence.

steps: Vec<String>

Registry ids of the steps that ran, in order. A report that covers part of the suite must say which part: absent evidence is not evidence.

request_count: u32

results: Vec<ProbeResult>

verdict: Verdict

identity: Identity

billing: BillingAudit

channel: ChannelSignature

perf: PerfSummary

skipped: Vec<String>

Probes that never ran, and why — so “not tested” is never read as “passed”.

Implementations

impl Report

pub fn by_group(&self, g: Group) -> Vec<&ProbeResult>

pub fn count(&self, s: Status) -> usize

pub fn exit_code(&self) -> i32

Exit code contract: 0 clean, 1 failing score, 2 hard gate tripped.

Trait Implementations

impl Clone for Report

fn clone(&self) -> Report

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Report

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for Report

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for Report

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

where __S: Serializer,

Serialize this value into the given Serde serializer.