Enum PromptError 

#[non_exhaustive]
pub enum PromptError {
    CompletionError(CompletionError),
    MemoryError(MemoryError),
    MaxTurnsError {
        max_turns: usize,
        chat_history: Box<Vec<Message>>,
        prompt: Box<Message>,
    },
    PromptCancelled {
        chat_history: Vec<Message>,
        reason: String,
    },
    UnknownToolCall {
        tool_name: String,
        available_tools: Vec<String>,
        allowed_tools: Vec<String>,
        chat_history: Box<Vec<Message>>,
    },
}

Errors from classic agent prompting.

Variants (Non-exhaustive)

Non-exhaustive enums could have additional variants added in future. Therefore, when matching against variants of non-exhaustive enums, an extra wildcard arm must be added to account for any future variants.

CompletionError(CompletionError)

A provider completion failed.

MemoryError(MemoryError)

Conversation memory failed to load or persist history.

MaxTurnsError

The run exhausted its total model-call budget.

Fields

max_turns: usize

Configured total model-call budget.

chat_history: Box<Vec<Message>>

Canonical history available when the budget was exhausted.

prompt: Box<Message>

Prompt for the call that could not be dispatched.

PromptCancelled

A prompting loop was cancelled.

Fields

chat_history: Vec<Message>

Canonical history available at cancellation.

reason: String

Human-readable cancellation reason.

UnknownToolCall

The model attempted to call a tool unavailable for the current turn.

Fields

tool_name: String

Tool name emitted by the model.

available_tools: Vec<String>

Tools registered on the runtime.

allowed_tools: Vec<String>

Exact immutable set allowed for this turn.

chat_history: Box<Vec<Message>>

Canonical history available at failure.

Implementations

impl PromptError

pub fn provider_response_body(&self) -> Option<&str>

Returns the provider response body exposed by a wrapped completion error.

pub fn provider_response_json(&self) -> Result<Option<Value>, Error>

Parses a wrapped provider response body as JSON when present.

pub fn provider_response_status(&self) -> Option<StatusCode>

Returns the HTTP status exposed by a wrapped completion error.

Trait Implementations

impl Debug for PromptError

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

Formats the value using the given formatter.

impl Display for PromptError

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

Formats the value using the given formatter.

impl Error for PromptError

fn source(&self) -> Option<&(dyn Error + 'static)>

Returns the lower-level source of this error, if any.

fn description(&self) -> &str

👎Deprecated since 1.42.0:

use the Display impl or to_string()

fn cause(&self) -> Option<&dyn Error>

👎Deprecated since 1.33.0:

replaced by Error::source, which can support downcasting

fn provide<'a>(&'a self, request: &mut Request<'a>)

🔬This is a nightly-only experimental API. (error_generic_member_access)

Provides type-based access to context intended for error reports.

impl From<Box<PromptError>> for StreamingError

fn from(source: Box<PromptError>) -> Self

Converts to this type from the input type.

impl From<Box<PromptError>> for StructuredOutputError

fn from(source: Box<PromptError>) -> Self

Converts to this type from the input type.

impl From<CompletionError> for PromptError

fn from(source: CompletionError) -> Self

Converts to this type from the input type.

impl From<MemoryError> for PromptError

fn from(source: MemoryError) -> Self

Converts to this type from the input type.

impl From<PromptError> for ExtractionError

fn from(source: PromptError) -> Self

Converts to this type from the input type.

impl From<PromptError> for ScenarioError

Available on crate feature test-utils only.

fn from(source: PromptError) -> Self

Converts to this type from the input type.