Struct CompletionRequest 

pub struct CompletionRequest {
    pub system: Option<String>,
    pub messages: Vec<Message>,
    pub max_tokens: u32,
    pub model_override: Option<String>,
    pub schema: Option<Value>,
    pub tools: Option<Vec<ToolRequest>>,
    pub tool_choice: Option<ToolChoice>,
}

Request for a text completion from an LLM provider.

Fields

system: Option<String>

Optional system prompt. Sent before the conversation messages.

messages: Vec<Message>

Conversation messages in chronological order.

max_tokens: u32

Maximum number of tokens in the response.

model_override: Option<String>

Optional per-request model override.

None resolves to the client’s LlmClient::default_model at call time.

schema: Option<Value>

Optional JSON schema for structured output.

Passed through to the provider as-is. Phase 166 adds a typed complete::<T>() wrapper with schemars normalization on top of this field. With streaming + schema, tokens arrive as raw JSON fragments — callers must accumulate before parsing (Pitfall 1).

tools: Option<Vec<ToolRequest>>

Optional tool definitions for the tool-calling dispatch loop.

Each entry’s parameters_schema must be pre-normalized via schema::for_structured_output. Set by ToolRegistry::dispatch (D-14).

tool_choice: Option<ToolChoice>

Controls how the LLM selects a tool when tools is Some.

Trait Implementations

impl Clone for CompletionRequest

fn clone(&self) -> CompletionRequest

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for CompletionRequest

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

Formats the value using the given formatter.