Module transcript 

Conversation validation and repair, assistant-turn classification, and constructors for real or synthetic tool results.

use rig_core::{message::Message, transcript::validate_canonical};

validate_canonical(&[Message::user("Hello"), Message::assistant("Hi")])?;

Enums

TranscriptError

Why a history is not a canonical transcript. See validate_canonical.

Constants

TOOL_NOT_EXECUTED_DUE_TO_INVALID_PEER

The result every other call of a turn gets when one call was retried or skipped: none of the turn’s calls ran.

Functions

assistant_text_from_choice

The text parts of an assistant turn, concatenated.

invalid_arguments_feedback

What the model reads when its call to tool sent raw, arguments that are not a JSON object: the tool never ran, and the model calls it again.

invalid_call_feedback

The tool results answering a turn with an invalid call, in call order: feedback for the call invalid, TOOL_NOT_EXECUTED_DUE_TO_INVALID_PEER for every other call. Empty when content has no tool calls.

is_empty_assistant_turn

Whether a generated assistant turn is empty: every part is blank (AssistantContent::is_blank, the rule replay drops parts by). An empty turn must not enter history.

not_executed

The result of a call the run did not execute because a call beside it was invalid.

tool_result_message

Constructs the error result of a call that never ran, its text verbatim, such as recovery feedback or a skip reason. JSON-shaped text is not reinterpreted as structured or multimodal output.

tool_result_output

Shape a tool’s result as the tool result the model reads, without reparsing text. Anything but a success is an error result.

validate_canonical

Rejects consecutive assistant messages, unanswered tool-call IDs, and results without a pending call. Each pending ID must be answered once in the next user message, before another assistant message or the end of history. System messages reset the consecutive-assistant check but retain pending calls. Duplicate call IDs are treated as one pending ID.