Crate ai_agents_eval 

Declarative evaluation runner for YAML-defined agents.

Re-exports

pub use assertion::AssertionOutcome;

pub use assertion::AssertionResultDetail;

pub use assertion::evaluate_assertion;

pub use evidence::DisambiguationStatus;

pub use evidence::ToolExecutionRecord;

pub use evidence::ToolExecutionSource;

pub use evidence::TurnEvidence;

pub use fixtures::FixturesConfig;

pub use fixtures::LlmFixtureMode;

pub use fixtures::RecordingToolLog;

pub use judge::JudgeConfig;

pub use judge::JudgeInput;

pub use judge::JudgeResolver;

pub use judge::JudgeResult;

pub use judge::LLMJudge;

pub use output::write_outputs;

pub use redaction::RedactedString;

pub use reset::ResetOptions;

pub use reset::ResetProfile;

pub use runner::EvalRunner;

pub use runner::EvalRunnerOptions;

pub use suite::EvalResult;

pub use suite::EvalSettings;

pub use suite::EvalSuite;

pub use suite::ScenarioResult;

pub use suite::ScenarioStatus;

Modules

assertion

compatibility

evidence

fixtures

judge

metrics

output

redaction

reset

runner

suite

Enums

EvalError

Error type used by the evaluation crate.

Type Aliases

Result