Free, local-first evaluation dashboard for agent CLIs with cross-harness transcript collection, live trace intelligence, deterministic graders, and accuracy audits.

Self-Improvement-Plugin gives an agent memory-aware startup, safer tool use, session closeout, verification hooks, and fresh-context delegation. It is built for long agent sessions where the agent needs to remember prior work, avoid repeated mistakes, recover from drift, and capture lessons before the context disappears. WIP, Open-Source, Free