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 |
llms.txt
View on GitHub# Roundtable Consilium Skill
> A Claude Code Skill that convenes 6 independent top-tier-model subagents as adversarial advisors when the user (小陈) needs to think through a difficult decision. Inspired by Roman consilium and rejection of AI sycophancy. The user sits at the head of the table — AI does not decide.
## Core Capabilities
- Extract topic briefing from current conversation context
- Hand-shake briefing with user before dispatching subagents
- Parallel-dispatch 6 top-tier-model subagents with distinct adversarial personas
- Each subagent isolated (cannot see other seats) to prevent echo chamber
- Seat 7 (real user feedback) explicitly NOT roleplayed — user must ask a real person
- AI refuses to make the final decision on user's behalf
- Auto-write meeting record to Obsidian vault with frontmatter
## Six Seats
| # | Seat | Stance |
|---|------|--------|
| 1 | Socrates | Asks 5-7 prerequisite-piercing questions, no answers |
| 2 | Senior Peer | Zero-emotion code review, lists 3-5 hard problems |
| 3 | Biased Investor | Already not investing, gives 3 specific reasons why |
| 4 | Historian | Lists 3-5 past projects that died the same way |
| 5 | Future-Regretful Self | Letter from 2031 self exposing today's self-deception |
| 6 | Ally | Believes it can succeed, gives 3 specific upgrades to not die |
Seat 7 (real user) is explicitly NOT AI-played.
## Tech Stack
- Claude Code Skill (SKILL.md format; seat 4 web verification prefers a configured search MCP such as grok-search, degrades to built-in WebSearch/WebFetch)
- Agent tool with `model: opus` for parallel subagent dispatch
- File output to Obsidian vault at `~/obsidian/灵感库/决策记录/`
## Trigger Phrases
Chinese: `/圆桌`, `开个圆桌`, `开圆桌`, `圆桌一下`
English: `/roundtable`, `open the roundtable`
## Cost Profile
- Full meeting: ¥8-20 per session (6 top-tier-model subagents)
- Single seat test: ¥1-2
- Dry-run with Haiku 4.5: ¥0.5
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 |