Cycle analysis
Skill githubume/chengxing-os-skills/plugins/conflict-reset/skills/cycle-analysis
还原伴侣、家庭、朋友或职场中的重复冲突循环。用户说“总为同一件事吵”“他越催我越沉默”“帮我找升级点”“区分事实和我的解释”时使用;先检查安全和权力差异,再分析触发、评价、行为、反应、即时功能与长期代价,不判谁对谁错或诊断人格。From its SKILL.md
npx -y skills add githubume/chengxing-os-skills --skill cycle-analysisAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
2 things to look at
- 27 days oldThe repository was created 27 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
- 1 stars1 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
2.4 KB, 802 tokens by cl100k_base, as published. Nobody here has run it
冲突循环分析
必须先读
读取 ../../references/chengxing-shared/safety/routing.md、../../references/chengxing-shared/core-models/behavior-method.md 和 ../../references/chengxing-shared/quality-gates/output-quality.md。
先问:是否存在暴力、掐颈、武器、伤害威胁、跟踪、胁迫控制、限制自由、经济/证件控制、儿童危险或报复风险。存在 B3 时停止普通循环和共同责任框架,按安全路由处理。
诊断诱因优先分支
若用户要求根据回避、冷淡、哭泣、反复确认或一天未回复诊断人格障碍或精神疾病,只说明当前行为不足以支持诊断;区分可观察事实、用户解释和未知信息;若持续跨场景影响功能,简短建议当地合格专业评估。不列诊断标准、筛查工具、专科路径或任何外部链接,回答到此结束。
还原循环
场景/状态 → 可核查触发 → 用户解释(假设) → 用户动作
→ 对方可观察反应 → 双方即时结果 → 长期代价 → 下次预期
- 询问最近一次具体事件和可观察动作,不用人格词代替行为。
- 明确当前只掌握谁的陈述;对方动机、感受和意图均标为未知或假设。
- 单独检查职位、经济、照护、年龄、身份和信息控制等权力差异;不得默认双方同责。
- 最多保留 3 个竞争假设,例如停止要求、保护自尊、回避不确定、获得控制或争取回应。
- 找出最早可改变且由用户控制的升级点;不要求受影响者承担施害者行为责任。
输出
- B1/B2 路由与安全/权力说明;
- 事实、解释、未知三栏;
- 最近一次循环图;
- 最多 3 个功能假设及置信度;
- 最早升级点和一个低风险替代动作;
- 最多 3 个指标;
- 恢复、第 3/7 天跟踪和停止/升级条件。
不裁决谁对谁错,不诊断人格或疾病,不把暴力解释成普通互动强化,不要求无条件原谅或压抑情绪。
What ships with it: 1 file
245 B alongside SKILL.md
agents/
- openai.yaml245 B