Enum FinishReason 

#[non_exhaustive]
pub enum FinishReason {
    NoMoreToolCalls,
    BudgetExceeded,
    ProviderStop(String),
    Stuck {
        repeated_call: String,
        repeats: usize,
    },
    TranscriptLimit {
        chars: usize,
        limit: usize,
    },
    PlanPending,
    Cancelled,
    PermissionDenialLimit,
}

Why the agent’s run terminated.

Variants

• NoMoreToolCalls: Model produced a response without tool calls (natural completion).
• BudgetExceeded: Ran out of steps (hit max_steps). Agent likely unfinished.
• ProviderStop(reason): LLM provider stopped unexpectedly. reason may be “length” (truncated by token limit), “stop”/“end_turn”, or a provider-specific code.
• Stuck: Agent got stuck calling the same tool repeatedly with the same arguments. repeated_call is the tool name, repeats is how many times before stopping.
• TranscriptLimit: Transcript size hit max_transcript_chars hard limit before compaction could reduce it further. Agent cannot continue. chars is final size, limit is the configured maximum.

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.

NoMoreToolCalls

Model generated final response without requesting more tools.

BudgetExceeded

Agent exceeded the maximum number of steps allowed.

ProviderStop(String)

LLM provider stopped with a specific reason or status code.

Stuck

Agent detected repeated identical tool calls (stuck loop).

Fields

repeated_call: String

repeats: usize

TranscriptLimit

Transcript size exceeded hard limit and cannot be reduced further.

Fields

chars: usize

limit: usize

PlanPending

Agent proposed a plan (PlanFirst mode) and is waiting for confirmation.

Cancelled

Agent was cancelled by a shutdown signal (SIGINT/SIGTERM).

PermissionDenialLimit

The auto permission classifier reached its denial limit (3 consecutive or 10 total denials). All subsequent tool calls are blocked to prevent denial loops.

Trait Implementations

impl Clone for FinishReason

fn clone(&self) -> FinishReason

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for FinishReason

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for FinishReason

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Eq for FinishReason

impl PartialEq for FinishReason

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

Tests for self and other values to be equal, and is used by ==.

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

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Serialize for FinishReason

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.

impl StructuralPartialEq for FinishReason