Enum AgentEvent 

#[non_exhaustive]
pub enum AgentEvent {
    Ready {
        session_id: String,
        model: Option<String>,
    },
    TextChunk {
        msg_id: String,
        text: String,
        channel: TextChannel,
    },
    Thought {
        msg_id: String,
        text: String,
    },
    ToolCallStart {
        call_id: String,
        name: String,
        input: Value,
    },
    ToolCallEnd {
        call_id: String,
        output: String,
        is_error: bool,
    },
    Plan {
        entries: Vec<PlanEntry>,
    },
    AskUser {
        ask_id: String,
        prompt: String,
        kind: AskKind,
        options: Vec<AskOption>,
    },
    PermissionRequest {
        req_id: String,
        tool: String,
        intent: Value,
        scope: PermissionScope,
    },
    Done {
        stop_reason: StopReason,
        usage: Usage,
    },
    Error {
        code: String,
        message: String,
    },
}

A streaming event from the agent. Subset of spec §7.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

Ready

Session has been initialized and is ready to accept prompts.

Fields

session_id: String

model: Option<String>

TextChunk

Streaming text chunk from the assistant.

Fields

msg_id: String

text: String

channel: TextChannel

Thought

Agent’s internal reasoning (when exposed by the model).

Fields

msg_id: String

text: String

ToolCallStart

Agent started a tool call.

Fields

call_id: String

name: String

input: Value

ToolCallEnd

Agent’s tool call returned a result.

Fields

call_id: String

output: String

is_error: bool

Plan

Agent published an updated plan (full state, replaces previous).

Fields

entries: Vec<PlanEntry>

AskUser

Agent asks the user a structured question.

Fields

ask_id: String

prompt: String

kind: AskKind

options: Vec<AskOption>

PermissionRequest

Agent requests permission for a security-sensitive action.

Fields

req_id: String

tool: String

intent: Value

scope: PermissionScope

Done

Turn complete. Usage stats included.

Fields

stop_reason: StopReason

usage: Usage

Error

Non-fatal error during the turn.

Fields

code: String

message: String

Trait Implementations

impl Clone for AgentEvent

fn clone(&self) -> AgentEvent

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AgentEvent

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

Formats the value using the given formatter.