Struct Completion 

pub struct Completion {
    pub content: Vec<ContentBlock>,
    pub usage: Usage,
    pub stop: StopReason,
}

One normalized completion: the parsed result of a single model call (ADR-0007).

This is not a serde mirror of any wire (Anthropic returns interleaved content blocks, OpenAI returns split fields); each wire parses its raw response into this shape. Following Grok Build’s normalized response, the assistant turn is an ordered list of ContentBlocks (Text / Thinking{signature} / ToolUse, in order) rather than pre-split text + tool_calls — so nothing is lost and the engine can append content straight into an assistant Message (ADR-0013), preserving thinking blocks and their signatures for replay.

Fields

content: Vec<ContentBlock>

The assistant turn’s content blocks, in order.

usage: Usage

Token accounting parsed from the terminal usage event.

stop: StopReason

Why the model stopped.

Implementations

impl Completion

pub fn tool_uses(&self) -> impl Iterator<Item = &ContentBlock>

The tool_use blocks the model emitted, in order.

pub fn has_tool_calls(&self) -> bool

Whether this completion asked to call any tools.

pub fn text(&self) -> Option<String>

The concatenated text of all text blocks, if any.

Trait Implementations

impl Clone for Completion

fn clone(&self) -> Completion

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Completion

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

Formats the value using the given formatter.

impl PartialEq for Completion

fn eq(&self, other: &Completion) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Completion