factormad-debate-factor-mining

Run a self-contained FactorMAD-style LLM multi-agent debate workflow for mining interpretable code-based stock alpha factors from daily OHLCV and point-in-time fundamental data. Use when an agent needs to generate, debate, validate, score, and export OHLCV, fundamental, or hybrid candidate alpha factor code for quantitative research.

跳到安装

仓库: XinAloha/skills
最近来源活动: 2026年8月22日 14:54
检测到的 SKILL.md 语言: 英语
星标: 3
分支: 0

安装方式

默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

直接命令不会经过审查 Prompt；运行前请先检查来源。

npx skills add https://github.com/XinAloha/skills --skill factormad-debate-factor-mining

命令会保持在同一行。复制前请横向滚动并检查完整内容。

想先保存到本地？可下载 SkillsMP 当前能够提供的文件。

检查来源文件

决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。

文件资源管理器

35 个文件

SKILL.md

来源说明 · 只读预览

返回搜索返回安装

backtest

Use when an agent needs a standard cross-sectional long-only backtest protocol with T+1 open execution, top-bucket equal weighting, fees, limit-up or limit-down exclusions, benchmark comparison, NAV curves, drawdown, IC, and diagnostic charts.

2026-08-22

brinson-performance-attribution

Brinson-Fachler / BHB sector attribution with interaction, concentration (HHI), top contributors, and optional Carino multi-period linking. Use to explain active return as allocation vs selection.

2026-08-22

factor-backtest

Run and explain cross-sectional trading-factor backtests on schema-compatible factor signals and market data. Use when Codex needs to validate backtest inputs; calculate returns, NAV, drawdown, holdings, turnover, benchmark or hedged performance; generate a PDF diagnostic report; or explicitly verify the checked-in test fixture. Also use for Chinese requests such as 因子回测、收益验证、净值、回撤、持仓、换手、对冲收益或回测报告.

2026-08-22

查看这个仓库的全部 Skills

name	factormad-debate-factor-mining
description	Run a self-contained FactorMAD-style LLM multi-agent debate workflow for mining interpretable code-based stock alpha factors from daily OHLCV and point-in-time fundamental data. Use when an agent needs to generate, debate, validate, score, and export OHLCV, fundamental, or hybrid candidate alpha factor code for quantitative research.
quantSkills	{"organization":"https://github.com/quantskills","repository":"quantskills/skill-factormad-debate-factor-mining","repository_url":"https://github.com/quantskills/skill-factormad-debate-factor-mining","project_type":"skill","collection":"factormad","license":"GPL-3.0-only","category":"tooling","tags":["factormad","alpha-factor-mining","multi-agent","debate","factor-debug"],"platforms":["codex"],"language":"zh-en","status":"active","validation_level":"runnable","maintainer_type":"community","requires":[],"summary_zh":"使用 FactorMAD 风格的 LLM 多智能体辩论流程从日频 OHLCV 与时点一致基本面数据中挖掘代码型股票 Alpha 因子。","summary_en":"Mine interpretable code-based stock alpha factors from daily OHLCV and point-in-time fundamental data with a FactorMAD-style LLM debate workflow."}

{
  "version": 1,
  "task": {
    "placeholder": "说明 OHLCV 数据、候选因子约束和实验目标；请上传或指明输入文件",
    "required": true
  },
  "fields": [
    {
      "key": "run_mode",
      "label": "运行模式",
      "type": "select",
      "default": "dry_run",
      "help": "建议先预检安装、路径和产物，再进行真实辩论",
      "options": [
        { "value": "dry_run", "label": "预检（不调用 LLM）" },
        { "value": "debate", "label": "真实多智能体辩论" }
      ]
    },
    {
      "key": "focus",
      "label": "挖掘关注点",
      "type": "textarea",
      "placeholder": "例如：量价背离、短期反转、可解释性或低换手"
    }
  ],
  "prompt_template": "{{#task}}任务与材料：\n{{task}}\n\n{{/task}}{{#attachments}}用户上传的材料（已放入工作区）：\n{{attachments}}\n\n{{/attachments}}按 FactorMAD 工作流以 {{run_mode}} 模式处理用户提供的 OHLCV 数据，{{#focus}}重点关注：{{focus}}；{{/focus}}组织多智能体提出、批判、校验、评分并导出可解释的 Python 股票 Alpha 候选代码，保留辩论轮次与接受因子产物，并明确内部 IC 证据不能替代独立样本外和组合验证，输出中文报告。"
}