Enum ExecutionEvent 

pub enum ExecutionEvent {
    Started {
        execution_id: String,
        at: DateTime<Utc>,
    },
    NodeStarted {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        iteration: u32,
    },
    NodeSucceeded {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        output: Value,
        iteration: u32,
    },
    NodeFailed {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        error: String,
        iteration: u32,
    },
    NodeSkipped {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        reason: String,
        iteration: u32,
    },
    NodeLog {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        stream: LogStream,
        line: String,
    },
    ToolCall {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        name: String,
        summary: String,
    },
    IterationCapped {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        visits: u32,
    },
    Paused {
        execution_id: String,
        at: DateTime<Utc>,
    },
    Resumed {
        execution_id: String,
        at: DateTime<Utc>,
    },
    AwaitingConfirmation {
        execution_id: String,
        node_id: String,
        action: String,
        at: DateTime<Utc>,
    },
    AiInvocation {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        provider: String,
        model: String,
        input: String,
        contract_version: Option<String>,
    },
    AiRoutingDecision {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        decision: String,
        confidence: Option<f64>,
        threshold: String,
        contract_version: String,
    },
    AiReviewRequired {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        primary_output: String,
        confidence: f64,
        reasoning: Option<String>,
    },
    AiReviewResolved {
        execution_id: String,
        node_id: String,
        at: DateTime<Utc>,
        approved: bool,
    },
    Done {
        execution_id: String,
        at: DateTime<Utc>,
        status: String,
    },
}

Variants

Started

Fields

execution_id: String

at: DateTime<Utc>

NodeStarted

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

iteration: u32

0-based execution count for this node. Always 0 on the acyclic path; increments when a bounded loop re-enters the node. Defaults so events serialized before iteration support deserialize cleanly.

NodeSucceeded

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

output: Value

iteration: u32

NodeFailed

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

error: String

iteration: u32

NodeSkipped

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

reason: String

iteration: u32

NodeLog

Incremental log line emitted by streaming adapters (the shell adapter is the first; future adapters can stream too). The Tauri side forwards these as flow:execution payloads which the Log Panel renders live.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

stream: LogStream

line: String

ToolCall

A model tool call dispatched inside an AI node’s in-node tool-use loop (an ai node bound to adapters). Emitted by the tool dispatcher just before the tool runs, so the UI shows live “editing X… running tests…” granularity mid-turn - individual tool calls otherwise produce no event, only node-level Started/Succeeded/Failed. UX/diagnostic layer like ExecutionEvent::NodeLog: not persisted to the run store.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

name: String

Full tool name, e.g. fs__edit-file.

summary: String

Human-readable one-line summary, e.g. edit-file src/foo.rs.

IterationCapped

A bounded loop hit the per-node visit cap and the node will not be scheduled again. Diagnostic only - the run finishes with partial status rather than aborting, so prior node outputs are preserved.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

visits: u32

Paused

The run reached a node-boundary checkpoint while paused and is now suspended until resumed or cancelled. Diagnostic - lets the UI confirm that an in-flight node finished and the run is truly idle.

Fields

execution_id: String

at: DateTime<Utc>

Resumed

A paused run was released and scheduling continues.

Fields

execution_id: String

at: DateTime<Utc>

AwaitingConfirmation

A node about to run performs a destructive operation and the per-step confirmation gate is enabled. The run self-pauses here until the user confirms (resume) or cancels (stop). action is a human-readable summary of what the node would do (e.g. “Delete file: build/”).

Fields

execution_id: String

node_id: String

action: String

at: DateTime<Utc>

AiInvocation

An AI node is invoking its model. Records the model input as a system event (RAO constraint 6.1). input is the post-sanitizer prompt; for cloud nodes without auditContent it is a short preview, matching the privacy posture of NodeSucceeded.output.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

provider: String

model: String

input: String

contract_version: Option<String>

Set when the node is contract-bound.

AiRoutingDecision

The engine routed a contract-bound AI output (RAO constraint 5/6). decision is auto_approve / human_review / suppress / contract_violation; threshold names the rule that applied, with its configured value, so the trail shows why the route was taken.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

decision: String

confidence: Option<f64>

threshold: String

contract_version: String

AiReviewRequired

A contract-bound AI output landed in the human review band (or the model set escalate). The run self-pauses until the gate is resolved (resolve_ai_review) or the run is cancelled.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

primary_output: String

confidence: f64

reasoning: Option<String>

AiReviewResolved

A human resolved an AI review gate. Recorded as a system event (RAO checklist 6.4): human decisions are execution events.

Fields

execution_id: String

node_id: String

at: DateTime<Utc>

approved: bool

Done

Fields

execution_id: String

at: DateTime<Utc>

status: String

Trait Implementations

impl Clone for ExecutionEvent

fn clone(&self) -> ExecutionEvent

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ExecutionEvent

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ExecutionEvent

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ExecutionEvent

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

where __S: Serializer,

Serialize this value into the given Serde serializer.