---
title: "empirical-macro-skills"
description: "Host-neutral Agent Skills for auditable empirical macro research with Agent Plan 专业数据集 integration."
canonical: https://agentpluginsdirectory.com/plugins/empirical-macro-skills
last-updated: 2026-10-07
---

# empirical-macro-skills
Host-neutral Agent Skills for auditable empirical macro research with Agent Plan 专业数据集 integration.
- Slug: empirical-macro-skills
- Publisher: Empirical Macro Skills contributors
- Repository: https://github.com/3494036618-eng/empirical-macro-skills
- Manifest: plugin.json
- Version: 0.2.0-beta
- License: Apache-2.0
- Category (editorial): research
- Skills: 6 (empirical-macro, macro-data, research-design, research-synthesis, robustness-audit, time-series-dynamics)
- MCP servers: 0
- Stars: 1
- Repository created: 2026-08-20
- Repository last pushed: 2026-08-24
- Publisher type: User
- Listing: https://agentpluginsdirectory.com/plugins/empirical-macro-skills
- Schema: https://agent-plugins.org/schemas/1.0.0/plugin.schema.json

## What empirical-macro-skills does, in the publisher's words

> 基于 Agent Plan 模型、专业数据集和豆包搜索，构建可审计、可复现、会主动 > 拒绝越界结论的宏观经济实证研究 Skill。

本项目是一套面向宏观经济实证研究的 Agent Skill 套件，覆盖研究设计、宏观数据 构建、动态分析、稳健性审计和研究综合。它通过结构化合同、确定性计算和可追溯 Artifact，把自然语言研究问题转化为可审计、可复现并受结论边界约束的研究流程。

Agent Plan 提供推荐的完整能力组合：

- 把模糊想法变成可检验、可审查的研究设计；
- 从专业数据集中筛选实体、指标、频率、时期和口径完全匹配的数据；
- 区分描述、关联、预测和因果，不让结论超过证据；
- 保存来源、参数、替代规格、失败记录和 SHA-256 校验值；
- 交付数据、表格、图形、报告和复现材料。
- Node.js 20 或更高版本（使用 npx 时）

From the project README, punctuation lightly normalized. Full text: https://raw.githubusercontent.com/3494036618-eng/empirical-macro-skills/HEAD/README.md

## Skills

- empirical-macro: Invoke as the mandatory first entry point when starting or resuming end-to-end empirical-macro research on monetary-policy shocks, inflation responses, data, robustness, or synthesis.
- macro-data: Use when audited macro-data preparation is the complete single-stage task and request JSON is validated; for end-to-end or multi-stage research, invoke empirical-macro first.
- research-design: Use when empirical-macro research design is the complete single-stage task; for end-to-end or multi-stage research, invoke empirical-macro first.
- research-synthesis: Use when synthesis of validated design, data, estimator, and robustness bundles is the complete single-stage task; for end-to-end or multi-stage research, invoke empirical-macro first.
- robustness-audit: Use when robustness auditing a validated estimator bundle is the complete single-stage task; for end-to-end or multi-stage research, invoke empirical-macro first.
- time-series-dynamics: Use when dynamic-path estimation is the complete single-stage task and upstream Artifacts are validated; for end-to-end or multi-stage research, invoke empirical-macro first.

Descriptions come from the frontmatter of each SKILL.md, punctuation lightly normalized.