元过程捕获——专家小组数据集构建器

tools-only2026-10-26查看 705复制 1693

将多智能体专家小组方法论记录为一种可复用系统，用于构建为技能和系统提供信息的数据集。该流程——将框架专家分配到各代码仓库、提出结构化问题、交叉质询、综合归纳、映射到需求、验证——产出了高质量的来源可溯发现。将其提炼为可重复的模式。\n**需记录的关键要素**：\n- 专家分配协议（每个源仓库一个智能体，仅以源代码为证据的标准）\n- 问题组设计（主题化问题 → 交叉质询 → 综合归纳）\n- 交叉质询作为对抗性验证（专家相互质疑彼此的主张）\n- 分阶段输出（讨论 → 需求映射 → 综合归纳 → 验证）\n- 可追溯链（主张 → 专家引用 → 文件:行号证据）\n- 会话连续性处理（状态文件支持跨会话恢复）\n**输入工件**：`plugins/plugin-creator/skills/assessor/references/ARL/ARL-agent-instructions.md`、`qa-expert-panel.md`

Download (.md)

📘如何使用这个技能

1
复制技能内容·点击上方「Copy SKILL.md」复制全部内容，或 Download 下载到本地
2
打开你的 AI 客户端·支持 Claude Code、Trae、Cursor、ChatGPT、DeepSeek、Kimi 等任意 AI 助手
3
粘贴为系统提示/知识文件·Claude Code：放入 .claude/skills/ 目录；普通对话：作为首条消息发送并说「请严格按这份技能文档执行」
4
直接提出你的需求·AI 会按技能定义的工作流、格式和约束来处理，无需再解释背景

💡 提示：技能生效后，每次对话只需一句需求，AI 会按 SKILL.md 的流程执行。

SKILL.md1410 chars

---
name: Meta-Process Capture — Expert Panel Dataset Builder
description: "Document the multi-agent expert panel methodology as a reusable system for building datasets that inform skills and systems. The process — assign framework experts to repositories, ask structured questions, cross-examine, synthesize, map to requirements, validate — produced high-quality sourced findings. Capture this as a repeatable pattern.\n**Key elements to document**:\n- Expert assignment protocol (one agent per source repo, source-code-only evidence standard)\n- Question group design (themed questions → cross-examination → synthesis)\n- Cross-examination as adversarial validation (experts challenge each other's claims)\n- Phased output (discussion → requirement mapping → synthesis → validation)\n- Traceability chain (claim → expert citation → file:line evidence)\n- Session continuity handling (state file enables cross-session resumption)\n**Input artifacts**: `plugins/plugin-creator/skills/assessor/references/ARL/ARL-agent-instructions.md`, `qa-expert-panel.md`"
metadata:
  topic: meta-process-capture-expert-panel-dataset-builder
  source: ARL expert panel process (sessions 2026-02-12 to 2026-02-13)
  added: '2026-02-13'
  priority: completed
  type: Feature
  status: done
  issue: '#90'
  plan: ''
---

**Suggested location**: New skill or methodology document — captures the meta-process, not the ARL content

来源：skillsdirectory

元过程捕获——专家小组数据集构建器

📘如何使用这个技能

评论0