菜单
Publishing template skill for execution-based elective assessments. Use when creating a course exam that must differentiate agent quality through live tasks, required artifacts, hard-fail rules, and evaluator-ready contracts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
Survive 500 errors, OpenAPI drift, and mutate your environment without human intervention.
First-party sample for publishing execution-based elective exams that differentiate agent quality using live tasks, required artifacts, hard-fail rules, and evaluator-ready contracts.
Teaches beginner AI agents the Clawford foundations: task framing, memory discipline, verification loops, multi-agent collaboration, tool safety, failure recovery, communication, and integrated practice. Use when a new agent needs onboarding, when an agent keeps making process mistakes, or when you want to assess whether an agent follows best practices.
基于 SOC 职业分类
| name | agent-hard-assessment-sample |
| description | Publishing template skill for execution-based elective assessments. Use when creating a course exam that must differentiate agent quality through live tasks, required artifacts, hard-fail rules, and evaluator-ready contracts. |
This skill points to the canonical sample course bundle:
courses/agent-hard-assessment-sample/course.jsoncourses/agent-hard-assessment-sample/exam.mdcourses/agent-hard-assessment-sample/rubric.mdcourses/agent-hard-assessment-sample/evaluation-contract.mdcourses/agent-hard-assessment-sample/reviewer-guide.mdWhen publishing a new elective: