< Summary

Information
Class: NexusLabs.Needlr.AgentFramework.Evaluation.EvaluationInputs
Assembly: NexusLabs.Needlr.AgentFramework.Evaluation
File(s): /home/runner/work/needlr/needlr/src/NexusLabs.Needlr.AgentFramework.Evaluation/EvaluationInputs.cs
Line coverage
100%
Covered lines: 2
Uncovered lines: 0
Coverable lines: 2
Total lines: 22
Line coverage: 100%
Branch coverage
N/A
Covered branches: 0
Total branches: 0
Branch coverage: N/A
Method coverage

Feature is only available for sponsors

Upgrade to PRO version

Metrics

MethodBranch coverage Crap Score Cyclomatic complexity Line coverage
get_Messages()100%11100%
get_ModelResponse()100%11100%

File(s)

/home/runner/work/needlr/needlr/src/NexusLabs.Needlr.AgentFramework.Evaluation/EvaluationInputs.cs

#LineLine coverage
 1using Microsoft.Extensions.AI;
 2
 3namespace NexusLabs.Needlr.AgentFramework.Evaluation;
 4
 5/// <summary>
 6/// Inputs shaped for <c>Microsoft.Extensions.AI.Evaluation</c> evaluators, derived from a
 7/// captured agent run. Consumers pass <see cref="Messages"/> and <see cref="ModelResponse"/>
 8/// to <c>IEvaluator.EvaluateAsync</c> (or to a <c>ScenarioRun</c> obtained via
 9/// <c>ReportingConfiguration.CreateScenarioRunAsync</c>).
 10/// </summary>
 11/// <param name="Messages">
 12/// The conversation messages that were sent to the agent, exactly as captured at the
 13/// agent-run boundary. Safe to replay without re-invoking the model.
 14/// </param>
 15/// <param name="ModelResponse">
 16/// The agent's aggregated response materialized as a <see cref="ChatResponse"/>. When
 17/// the captured run produced no output, this is an empty assistant response so evaluators
 18/// still receive a non-null value.
 19/// </param>
 20public readonly record struct EvaluationInputs(
 621    IReadOnlyList<ChatMessage> Messages,
 1722    ChatResponse ModelResponse);

Methods/Properties

get_Messages()
get_ModelResponse()