chenyuxiaojin/xiaochen-skills / plugins/cyxj-roundtable/skills/cyxj-roundtable
cyxj-roundtable
Verified圆桌会议 skill。当你想不明白一件事时,召集 6 个独立 subagent(当前最强对话模型档) 扮演 6 个有立场的角色(苏格拉底/严苛同行/带偏见投资人/历史学家/5 年后后悔的你/同盟者), 从 6 个不会互相迁就的视角同时给出意见。主位坐你,AI 不替你拍板。 会议结束自动落地到 Obsidian 决策记录。 触发方式:/圆桌、"开个圆桌"、"开圆桌"、"圆桌一下"。 Roundtable consilium. Convenes 6 independent top-tier-model subagents with adversarial perspectives to help you decide. You sit at the head. AI does not decide for you. Trigger: /roundtable, "open the roundtable", "/圆桌"
| File | Kind | Size |
|---|---|---|
| README.md | markdown | 1.6 KB |
| SKILL.md | skill md | 7.6 KB |
| llms.txt | markdown | 1.9 KB |
| seats/01-socrates.md | markdown | 950 B |
| seats/02-peer.md | markdown | 1005 B |
| seats/03-investor.md | markdown | 984 B |
| seats/04-historian.md | markdown | 2.9 KB |
| seats/05-future-self.md | markdown | 1.1 KB |
| seats/06-ally.md | markdown | 1.1 KB |
| templates/briefing.md | markdown | 463 B |
| templates/record.md | markdown | 1.0 KB |
README.md
View on GitHub# cyxj-roundtable:圆桌会议 skill
## 是什么
当你对一件事想不明白时,在 Claude Code 对话里召集"圆桌"。
6 个独立 subagent(当前 opus 档模型) 同时扮演 6 个有立场的角色:
- 🪑 苏格拉底(追问者)
- 🪑 严苛同行(零情绪 Code Review)
- 🪑 带偏见投资人(已经决定不投,找 3 个理由)
- 🪑 历史学家(摆过去同类项目的尸体)
- 🪑 5 年后后悔的你(戳穿当年的自欺)
- 🪑 同盟者(不夸你,帮你不死)
第 7 席留给真人。主位坐你。
## 为什么这样设计
来自 2026-05-18 日记的思考——AI 默认是"邹忌三镜"里的私我/畏我/有求于我,
廉价地说"你方案非常棒"。让 AI 真正反驳自己,需要刻意搭一个会反驳你的对话环境。
6 席用独立 subagent 实现,避免单一模型自我迁就的回音壁。
## 怎么用
在已经聊了一阵某件事想不明白时,直接说:
```
开个圆桌
```
或:`/圆桌`、`开圆桌`、`圆桌一下`、`/roundtable`
接下来:
1. Claude 从上下文提炼"议题简报"给你确认
2. 你说"开会" → 6 席并行思考 1-2 分钟
3. 6 席发言后,Claude 输出"建议去问哪个真人"(第 7 席)
4. 提炼 6 席共识和分歧 → 等你拍板
5. 自动写决策记录到 `~/obsidian/灵感库/决策记录/`
## 成本
- 单次完整开会:约 ¥8-20(6 个 subagent(当前 opus 档模型))
- 单席测试:约 ¥1-2
- Dry-run(Haiku 4.5):约 ¥0.5
## 设计文档
完整设计见 `../../docs/2026-05-20-roundtable-design.md`。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |