OutcomeVerifier class
Outcome Verifier - Detects success/failure of critical UI actions.
Uses a two-stage approach:
- Deterministic verification - Pattern matching for success/error signals
- LLM verification - Falls back to AI analysis when uncertain
Constructors
- OutcomeVerifier({required AiProvider provider, required AgentConfig config})
Properties
- config → AgentConfig
-
final
- hashCode → int
-
The hash code for this object.
no setterinherited
- provider → AiProvider
-
final
- runtimeType → Type
-
A representation of the runtime type of the object.
no setterinherited
Methods
-
getMaxFollowupSteps(
) → int - Get maximum followup steps for recovery attempts.
-
isCriticalAction(
VerificationAction action) → bool - Check if an action is critical (requires verification).
-
isEnabled(
) → bool - Check if verifier is enabled.
-
noSuchMethod(
Invocation invocation) → dynamic -
Invoked when a nonexistent method or property is accessed.
inherited
-
toString(
) → String -
A string representation of this object.
inherited
-
verify(
VerificationContext context) → Future< VerificationResult> - Verify action outcome using deterministic and LLM-based methods.
Operators
-
operator ==(
Object other) → bool -
The equality operator.
inherited