Struct RunOptions 

pub struct RunOptions {
    pub min_new: Option<u64>,
    pub min_new_errors: Option<u64>,
    pub if_stale_ms: Option<i64>,
    pub namespaces: Vec<String>,
    pub full_sweep: bool,
    pub triggering_actor: Option<String>,
}

Gating and scoping options for a run.

Fields

min_new: Option<u64>

min_new_errors: Option<u64>

if_stale_ms: Option<i64>

namespaces: Vec<String>

Optional global namespace filter (empty = all).

full_sweep: bool

Re-analyze the WHOLE memory this pass, not just grains since the last-run watermark (areev loop reflect). Dedup/cooldowns still suppress anything already queued, and the watermark still advances at the end — so a sweep is safe to run any time and later runs stay incremental. Mainly widens the watermark-sensitive inputs (tool-failure window, the non-parasitic LLM evidence bundle) to the full history.

triggering_actor: Option<String>

The principal that invoked this run. Recorded as co-creator on every non-Builtin (LLM / external-command) recommendation the run stores, so the review gate’s self-approval block also fires for whoever triggered the model that authored the finding — an LLM draft is authored via its trigger, unlike a deterministic finding, which is computed. None (a headless/scheduled run) records no co-creator.

Trait Implementations

impl Clone for RunOptions

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for RunOptions

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

Formats the value using the given formatter.

impl Default for RunOptions

fn default() -> Self

Returns the “default value” for a type.