Scenario Leaderboard

Skill-aligned scenarios: ES|QL, log investigation, and security triage. Compare baseline vs skill-augmented performance.

All0With Uplift Data0Skills Enabled0Baseline Only0
0 results

How Scenario Scoring Works

Each model runs 12 skill-aligned scenarios against real Elasticsearch (via Docker or Cloud). With --compare-skills, each scenario runs twice: once baseline (no skill context) and once with skills (Elastic Agent Skill injected into the prompt). The Skill Uplift shows the delta.

Loading...