Package dev.dokimos.core.evaluators.agents

package dev.dokimos.core.evaluators.agents

Related Packages

Package	Description
dev.dokimos.core.evaluators

Class	Description
AgentEvalCase	Typed entry point for building an EvalTestCase for the agent evaluators.
AgentEvalCase.Builder	Builder for constructing an EvalTestCase that targets the agent evaluators.
ArgMatchMode	Controls how a tool call's arguments are compared against the expected arguments.
ArgumentMatcher	Strategy for comparing a tool call's actual arguments against the expected arguments.
TaskCompletionEvaluator	Evaluates whether an AI agent completed the user's requested tasks.
TaskCompletionEvaluator.Builder	Builder for constructing the evaluator.
TolerantArgumentMatcher	Default ArgumentMatcher that compares arguments structurally with a few deliberate tolerances.
TolerantArgumentMatcher.Builder	Builder for constructing the matcher.
ToolArgumentHallucinationEvaluator	Uses a judge LLM to assess whether tool call argument values are factually grounded in the user's input and preceding tool call results.
ToolArgumentHallucinationEvaluator.Builder	Builder for constructing the evaluator.
ToolCallValidityEvaluator	Validates that tool calls are syntactically correct per their JSON schema definitions.
ToolCallValidityEvaluator.Builder	Builder for constructing the evaluator.
ToolCorrectnessEvaluator	Checks whether the agent used the expected set of tools.
ToolCorrectnessEvaluator.Builder	Builder for constructing the evaluator.
ToolCorrectnessEvaluator.MatchMode	The comparison mode for evaluating tool correctness.
ToolDescriptionReliabilityEvaluator	Evaluates tool description quality using a mix of rule-based checks and optional LLM checks.
ToolDescriptionReliabilityEvaluator.Builder	Builder for constructing the evaluator.
ToolEfficiencyEvaluator	Measures how efficiently an agent used its tools by detecting redundant calls.
ToolEfficiencyEvaluator.Builder	Builder for constructing the evaluator.
ToolErrorEvaluator	Detects tool execution failures by inspecting ToolCall.result().
ToolErrorEvaluator.Builder	Builder for constructing the evaluator.
ToolNameReliabilityEvaluator	Evaluates tool naming quality using a mix of rule-based checks and optional LLM checks.
ToolNameReliabilityEvaluator.Builder	Builder for constructing the evaluator.
ToolTrajectoryEvaluator	Evaluates an agent's tool-call trajectory against an expected trajectory using a selectable match mode.
ToolTrajectoryEvaluator.Builder	Builder for constructing the evaluator.
ToolTrajectoryEvaluator.MatchMode	The comparison mode for evaluating a trajectory.