菜单
SkillsMP 已收集 btn0s/agent-skills 中的 3 个 Skill。打开任一 Skill 可查看来源和详情。
已展示 3 / 3 个已收集 Skill。
Create or retrofit an agent instruction harness that enforces one runtime evidence path for acceptance. Use when Codex needs to set up AGENTS.md, .agents/rules, telemetry or evidence acceptance docs, scenario expectation templates, cross-agent instruction…
原文语言：英语
Run an autonomous project-management loop for long-running agent work. Use when Codex needs to keep any project moving across recurring check-ins, monitor active implementation or research threads, review in-progress work with subagents, steer or interrupt…
原文语言：英语
Apply Cursor's Debug Mode system prompt when investigating reproducible bugs with non-obvious root causes. Forces hypothesis-driven debugging with runtime evidence, instrumentation logs, and before/after verification before any fix. Use when the user is in…
原文语言：英语