Evaluation Methods for Agent Systems

Evaluate agent systems differently from traditional software because agents make dynamic decisions, are non-deterministic between runs, and often lack single correct answers. Build evaluation frameworks that account for these characteristics, provide actionable feedback, catch regressions, and validate that context engineering choices achieve intended effects.

When to Activate

• Testing agent performance systematically
• Validating context engineering choices
• Measuring improvements over time
• Catching regressions before deployment
• Building quality gates for agent pipelines
• Comparing different agent configurations
• Evaluating production systems continuously
• Building an LLM judge: direct scoring, pairwise comparison, rubric generation, calibration against human ratings, bias mitigation

Do not activate this skill for adjacent work owned by other skills:

• Debugging a specific context failure mode before measuring it: context-optimization (diagnose mode, references/degradation.md).
• Tool API contracts for evaluation tools: tool-design.