Quality
Evaluations in Optimizely Mark overview
Evaluations measure how well an agent in performs against your quality standards. Use evaluations to score agent outputs, catch unexpected behavior, and improve agent quality over time.
Quality tab overview
The Quality tab gives you automated quality scoring and runtime safeguards for every specialized agent execution in . Define your quality criteria, score each execution against them, and recover or block executions that deviate from expected behavior.
Configure Output Evaluation
Output Evaluation scores each specialized agent run against quality criteria, examples, and a baseline score you define. Use Output Evaluation to measure agent quality consistently as you iterate, without manually reviewing every run.
Configure Execution Guardrails
Execution Guardrails detect when a specialized agent run behaves unlike a normal run for your agent. Use Execution Guardrails to scale specialized agents safely across your organization without manually reviewing every execution.
Execution Advisor
Execution Advisor is the runtime intervention layer for specialized agents in . It works alongside Output Evaluation and Execution Guardrails to keep agent runs on course.
Preferred output examples
Preferred output examples are agent outputs you save as quality benchmarks for Output Evaluation. uses these examples to score future runs of the same agent against your defined standard.