Package com.tnsai.evaluation

package com.tnsai.evaluation

TnsAI Evaluation Framework - unified evaluation, benchmarking, and quality gating.

Architecture

• Layer 1 - Evaluators: Score agent responses on multiple dimensions
• Layer 2 - Benchmark Engine: Run test datasets, collect results, detect regressions
• Layer 3 - Reporting: Quality gates, trend analysis, aggregated reports

Since:

0.1.0

Related Packages

Package	Description
com.tnsai.evaluation.autoharness	Self-improving agent evaluation loop (auto-harness pattern).
com.tnsai.evaluation.benchmark	Layer 2: Benchmark engine for running test datasets, collecting results, and detecting regressions against saved baselines.
com.tnsai.evaluation.bridge
com.tnsai.evaluation.evaluators	Layer 1: Evaluators that score agent behavior on specific dimensions.
com.tnsai.evaluation.junit
com.tnsai.evaluation.model	Data model records for evaluation results, test cases, baselines, and configuration.
com.tnsai.evaluation.reporting	Layer 3: Quality gates, evaluation reports, and trend analysis.
com.tnsai.evaluation.spi	SPI interfaces for the evaluation framework.
com.tnsai.evaluation.util