Struct EngineConfig 

pub struct EngineConfig {

    pub tick_interval: Duration,
    pub sync_interval: Duration,
    pub webhook_health_check_interval: Option<Duration>,
    pub max_parallel: usize,
    pub stuck_timeout: u64,
    pub no_session_stuck_timeout: u64,
    pub in_review_no_session_stuck_timeout: u64,
    pub auto_create_followup_on_changes: bool,
    pub graceful_shutdown_timeout: Duration,
    pub silence_grace_period: u64,
    pub silence_cooldown: u64,
    pub upgrade_check_interval: u64,
    pub startup_failure_escalation_secs: u64,
}

Engine configuration.

Fields

tick_interval: Duration

Main tick interval

sync_interval: Duration

GitHub sync interval (cleanup, PR review, mentions)

webhook_health_check_interval: Option<Duration>

Webhook health check interval (seconds)

max_parallel: usize

Maximum parallel task executions

stuck_timeout: u64

Stuck task timeout for tasks with an active tmux session (seconds)

no_session_stuck_timeout: u64

Stuck task timeout for tasks with no active tmux session (seconds). Shorter than stuck_timeout because no session means the agent has already exited.

in_review_no_session_stuck_timeout: u64

Stuck task timeout for in_review tasks with no active tmux session (seconds). Longer than no_session_stuck_timeout because review agents exit their tmux session on normal completion before the result is delivered to the engine. Using a short no-session timeout for in_review tasks causes a race where a completed review result arrives after the stuck recovery has already reset the task to needs_review, discarding the review work.

auto_create_followup_on_changes: bool

Auto-create follow-up tasks when PR reviews request changes

graceful_shutdown_timeout: Duration

Graceful shutdown timeout — how long to wait for running agents before exiting.

silence_grace_period: u64

Grace period before silence detection kicks in (seconds). Allows time for agent startup, model download, API handshake.

silence_cooldown: u64

Cooldown duration for a model detected as silent (seconds).

upgrade_check_interval: u64

How often to check for a newer orch release and notify channels (seconds). Set to 0 to disable. Default: 3600 (1 hour).

startup_failure_escalation_secs: u64

Seconds of continuous startup failure (GitHub unreachable / auth errors) before the engine escalates from WARN to ERROR-level logging and sends a one-time push notification to configured channels. Set to 0 to disable. Default: 3600 (1 hour).

Implementations

impl EngineConfig

pub fn from_config() -> Self

Load engine configuration from config files, falling back to defaults.

Trait Implementations

impl Clone for EngineConfig

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Default for EngineConfig

fn default() -> Self

Returns the “default value” for a type.