'max-iterations' 검색 결과

Evaluate a Claude Managed Agent against a written rubric, a set of checkable criteria, and guide improvements within a fixed attempt limit. After a version passes, prepare separate evaluation cases to check performance beyond the examples used for revision.

한국어 요약은 준비 중입니다. 영어 요약을 표시하고 있습니다.

• Defining checkable criteria for agent outputs
• Improving an agent within a fixed attempt limit
• Testing a passing version on unseen cases