Agents And SkillsSkill

Claude Code self-improving skills loop with eval scoring

April 5, 2026Mike Futia

Mike Futia describes an internal Claude Code loop that repeatedly runs tests, scores outputs against an eval, rewrites the prompt, and keeps the best version, turning skills into iteratively optimized agents.

Self-improving Claude Code skills are f*cking ridiculous 🤯
One loop → 10 test runs, scored against an eval, prompt rewritten, retested, winner kept.
All inside Claude Code.
Mike Futia
coding agentsevalsclaudeanthropicclaude codecoding agents

See what authorities are saying right now

This finding is one of many signals tracked across Artificial Intelligence. The live feed updates every few hours with new authority voices, debates, and emerging ideas.

← Back to Artificial Intelligence