4-stage evaluation framework for testing Claude Code plugin component triggering. Validates skills, agents, and commands activate correctly via programmatic detection and LLM judgment.

Benchmark harness for A/B testing Claude Code plugins against OOLONG long-context reasoning tasks. Compare truncation vs RLM-RS recursive chunking strategies. Features Claude Code hooks integration, SQLite persistence, and comprehensive scoring aligned with the OOLONG paper methodology.

🚀 Automate the evaluation of Claude Code plugin components to ensure accurate triggering of skills, agents, commands, and hooks.