Struct GoldConfig 

pub struct GoldConfig {
    pub enabled: bool,
    pub auto_answer_enabled: bool,
    pub auto_continue_enabled: bool,
    pub model_name: Option<String>,
    pub goal: Option<String>,
    pub evaluation_prompt: Option<String>,
    pub max_output_tokens: u32,
    pub max_auto_continuations: u32,
    pub min_auto_continue_confidence: GoldConfidence,
}

Fields

enabled: bool

Master switch for Gold observe-only evaluation.

auto_answer_enabled: bool

Independent switch for Phase 2 low-risk auto-answer.

Kept separate from enabled so Phase 1 observe-only users do not implicitly opt into automatic clarification responses.

auto_continue_enabled: bool

Independent switch for Phase 3 server-side auto-continue.

Kept separate from both enabled and auto_answer_enabled so users can opt into terminal auto-resume explicitly without enabling other Gold automation behaviors.

model_name: Option<String>

Optional dedicated model for Gold evaluation. Falls back to fast model, then the main chat model when absent.

goal: Option<String>

The user’s goal for this session.

Unlike evaluation_prompt (which only tunes the judge), the goal is surfaced to the main executing agent as a persistent system-prompt block so it actively works toward it. The Gold evaluator also measures progress against this text.

evaluation_prompt: Option<String>

Optional custom prompt suffix appended to the built-in Gold evaluator prompt. This tunes the judge only; it does not set the goal.

max_output_tokens: u32

Output token limit for the Gold evaluator call.

max_auto_continuations: u32

Maximum number of automatic Gold continuations allowed per session.

min_auto_continue_confidence: GoldConfidence

Minimum evaluator confidence required before Gold auto-continues or auto-answers. Defaults to medium so the loop fires on reasonably confident verdicts rather than only high.

Implementations

impl GoldConfig

pub fn effective_goal(&self) -> Option<&str>

The session goal text, falling back to the legacy evaluation_prompt for sessions created before the dedicated goal field existed.

Returns None when neither field holds non-empty text.

Trait Implementations

impl Clone for GoldConfig

fn clone(&self) -> GoldConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for GoldConfig

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

Formats the value using the given formatter.

impl Default for GoldConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for GoldConfig

where GoldConfig: Default,

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for GoldConfig

impl PartialEq for GoldConfig

fn eq(&self, other: &GoldConfig) -> 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 GoldConfig

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

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for GoldConfig