Struct ReasonResult 

pub struct ReasonResult {

    pub native_counts: Option<NativeExecutionCounts>,
    pub success: bool,
    pub text: String,
    pub tool_calls: Vec<ToolCall>,
    pub has_tool_calls: bool,
    pub tool_definitions: Vec<ToolDefinition>,
    pub max_iterations: usize,
    pub error: Option<String>,
    pub user_facing_error: Option<UserFacingError>,
    pub error_disclosure: Option<ErrorDisclosure>,
    pub usage: Option<TokenUsage>,
    pub output_message_id: Option<MessageId>,
    pub time_to_first_token_ms: Option<u64>,
    pub response_id: Option<String>,
    pub finish_reason: Option<String>,
    pub locale: Option<String>,
    pub network_access: Option<NetworkAccessList>,
    pub parallel_tool_calls: Option<bool>,
    pub waiting_for_tool_results: bool,
}

Result of the ReasonAtom

Fields

native_counts: Option<NativeExecutionCounts>

success: bool

Whether the LLM call succeeded

text: String

Text response from the model

tool_calls: Vec<ToolCall>

Tool calls requested by the model

has_tool_calls: bool

Whether tool execution is needed

tool_definitions: Vec<ToolDefinition>

Tool definitions from applied capabilities (for tool execution)

max_iterations: usize

Maximum iterations configured for the agent

error: Option<String>

Error message if the call failed

user_facing_error: Option<UserFacingError>

Disclosed user-facing decision of the failure, already filtered through the resolved error-disclosure mode. Hosts must prefer this over re-classifying error/text strings so disclosure stays consistent.

error_disclosure: Option<ErrorDisclosure>

Error-disclosure mode that was applied to user_facing_error.

usage: Option<TokenUsage>

Token usage from the LLM call

output_message_id: Option<MessageId>

Assistant message emitted by output.message.completed for this generation.

time_to_first_token_ms: Option<u64>

Streaming latency for this LLM call, when available.

response_id: Option<String>

LLM provider’s response ID for chaining with previous_response_id

finish_reason: Option<String>

Raw provider finish reason for this generation.

locale: Option<String>

Resolved locale used for this turn’s prompt and backend-authored strings.

network_access: Option<NetworkAccessList>

Merged network access list for URL filtering in tools.

parallel_tool_calls: Option<bool>

Request-level parallel tool calling preference (EVE-598), carried from the resolved agent config into ActInput so the act scheduler can honor Some(false) (force serialize). None preserves the default schedule.

waiting_for_tool_results: bool

A remote tool loop (OpenAI Agents API) paused on a tool call; the turn parks (EVE-1124).

Trait Implementations

impl Clone for ReasonResult

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 ReasonResult

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

Formats the value using the given formatter.

impl Default for ReasonResult

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for ReasonResult

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

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for ReasonResult

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

where __S: Serializer,

Serialize this value into the given Serde serializer.