Struct AgentConfig 

pub struct AgentConfig {
    pub system_prompt: Option<String>,
    pub enable_thought: bool,
    pub reasoning: Option<ReasoningConfig>,
    pub language: Language,
    pub execution: ExecutionConfig,
    pub llm: LlmConfig,
    pub tool: ToolConfig,
    pub session: SessionConfig,
    pub safety: SafetyConfig,
}

Fields

system_prompt: Option<String>

enable_thought: bool

Controls whether to include the reasoning content in LLM responses.

Distinction from reasoning.enabled:

• enable_thought: controls whether the reasoning_content field is forwarded to consumers (i.e., “show the thinking process”)
• reasoning.enabled: controls whether the model’s extended thinking / reasoning mode is enabled (i.e., “let the model think deeply”)

Both are usually kept in sync, but can be controlled independently. For example, to enable deep thinking without showing the process, set enable_thought = false and reasoning.enabled = true.

reasoning: Option<ReasoningConfig>

Reasoning/thinking configuration that controls LLM reasoning behavior.

• enabled: whether to enable extended thinking mode (equivalent to builder’s enable_thinking)
• budget_tokens: thinking token budget cap
• effort: reasoning intensity/depth (semantics vary by provider)

language: Language

execution: ExecutionConfig

llm: LlmConfig

tool: ToolConfig

session: SessionConfig

safety: SafetyConfig

Implementations

impl AgentConfig

pub fn validate(&self) -> AgentResult<()>

Validate the configuration, returning an error for invalid values.

Trait Implementations

impl Clone for AgentConfig

fn clone(&self) -> AgentConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AgentConfig

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

Formats the value using the given formatter.

impl Default for AgentConfig

fn default() -> Self

Returns the “default value” for a type.