Struct ResultMessage 

pub struct ResultMessage {

    pub subtype: ResultSubtype,
    pub is_error: bool,
    pub duration_ms: u64,
    pub duration_api_ms: u64,
    pub ttft_ms: Option<u64>,
    pub ttft_stream_ms: Option<u64>,
    pub time_to_request_ms: Option<u64>,
    pub time_to_request_from_spawn_ms: Option<u64>,
    pub warm_spare_claimed: Option<bool>,
    pub time_origin_ms: Option<u64>,
    pub request_sent_wall_ms: Option<f64>,
    pub user_message_uuid: Option<String>,
    pub user_message_uuids: Vec<String>,
    pub queued_turn_count: Option<u64>,
    pub num_turns: i32,
    pub result: Option<String>,
    pub session_id: String,
    pub total_cost_usd: f64,
    pub usage: Option<UsageInfo>,
    pub permission_denials: Vec<PermissionDenial>,
    pub errors: Vec<String>,
    pub uuid: Option<String>,
    pub api_error_status: Option<u16>,
    pub stop_reason: Option<String>,
    pub terminal_reason: Option<String>,
    pub fast_mode_state: Option<String>,
    pub fast_mode_disabled_reason: Option<FastModeDisabledReason>,
    pub model_usage: Option<BTreeMap<String, ModelUsageEntry>>,
    pub subagent_stats: Option<SubagentStats>,
    pub structured_output: Option<Value>,
    pub deferred_tool_use: Option<DeferredToolUse>,
    pub origin: Option<MessageOrigin>,
}

Result message for completed queries

Fields

subtype: ResultSubtype

is_error: bool

duration_ms: u64

duration_api_ms: u64

ttft_ms: Option<u64>

Time to first token, in milliseconds.

ttft_stream_ms: Option<u64>

Time to first streamed token, in milliseconds.

time_to_request_ms: Option<u64>

Time from session start until the first request was issued, in milliseconds.

time_to_request_from_spawn_ms: Option<u64>

Time from spawning a worker/spare until the first request was issued, in milliseconds.

warm_spare_claimed: Option<bool>

Whether a warm spare process was claimed for this request.

time_origin_ms: Option<u64>

Epoch-ish timestamp origin used by CLI timing instrumentation.

request_sent_wall_ms: Option<f64>

Wall-clock epoch milliseconds when the API request was sent (fractional; from CLI timing instrumentation).

user_message_uuid: Option<String>

Wire uuid of the user message this result answers.

user_message_uuids: Vec<String>

Client uuids of every user message whose prompt this turn consumed, in consumption order — all members of a prompt batch the host merged into this one turn, plus any queued user message folded into the running turn between tool rounds. Always contains user_message_uuid; at most 64 entries. Absent on delivery-failure/zeroed results and from CLIs before 2.1.259 (fall back to user_message_uuid).

queued_turn_count: Option<u64>

User-initiated sends still waiting in the command queue when this result was produced. Greater than 0 means at least one more user turn follows without further input, barring cancellation. Absent on fatal startup results, on surfaces without a command queue, and from CLIs before 2.1.259.

num_turns: i32

result: Option<String>

session_id: String

total_cost_usd: f64

usage: Option<UsageInfo>

permission_denials: Vec<PermissionDenial>

Tools that were blocked due to permission denials during the session

errors: Vec<String>

Error messages when is_error is true.

Contains human-readable error strings (e.g., “No conversation found with session ID: …”). This allows typed access to error conditions without needing to serialize to JSON and search.

uuid: Option<String>

api_error_status: Option<u16>

HTTP status code when the result is an API error (e.g., 429, 500, 529)

stop_reason: Option<String>

Why generation stopped (e.g., end_turn, max_tokens)

terminal_reason: Option<String>

Why the session ended (e.g., “completed”)

fast_mode_state: Option<String>

Fast mode toggle state (e.g., “off”)

fast_mode_disabled_reason: Option<FastModeDisabledReason>

Why fast mode can’t serve right now. Absent when nothing blocks it. A paused-after-rate-limit run is not reported here; it rides fast_mode_state as "cooldown".

model_usage: Option<BTreeMap<String, ModelUsageEntry>>

Per-model cost breakdown, keyed by model name (e.g. "claude-opus-4-8").

subagent_stats: Option<SubagentStats>

Subagents started through the Agent tool in this session, as running totals. Cumulative like modelUsage: read the latest result rather than summing across results. Absent from CLIs before 2.1.239.

structured_output: Option<Value>

Structured-output payload returned by the model, when enabled.

deferred_tool_use: Option<DeferredToolUse>

Deferred tool-use termination payload.

origin: Option<MessageOrigin>

Provenance of the message/run.

Trait Implementations

impl Clone for ResultMessage

fn clone(&self) -> ResultMessage

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ResultMessage

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for ResultMessage

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ResultMessage

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

where __S: Serializer,

Serialize this value into the given Serde serializer.