Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/eddmpython/dartlab/macronpx skills add eddmpython/dartlab --skill macrogit clone --depth 1 https://github.com/eddmpython/dartlabWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.04570 |
| Opus 5 | $0.00000 | $0.02285 |
| Sonnet 5 | $0.00000 | $0.00914 |
| Haiku 4.5 | $0.00000 | $0.00457 |
Grade A, and why
macro scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 248 lines — stays where its author put it; the contents beside it link to each section on GitHub.
엔진 역할
macro는 회사가 아니라 시장/경제 환경을 읽는 L2 엔진이다. 경제 사이클, 재고, 기업집계, 교역, 전파 edge, 금리, 유동성, 위기, 자산, 심리, 예측, 시나리오, 전망 시뮬레이션, 종합을 6막 인과 구조로 해석한다.
단일 기업 수익성/현금흐름/가치평가는 analysis가 담당한다. macro는 그 기업이 놓인 외부 환경을 제공하고, 보고서 조합은 story가 담당한다.
공개 호출 방식
import dartlab
guide = dartlab.macro()
cycle = dartlab.macro("cycle", market="KR")
rates = dartlab.macro("금리", market="US")
scenario = dartlab.macro("시나리오", "2008 금융위기")
simulation = dartlab.macro("시뮬레이션", market="US") # BVAR 변수 팬(분위 경로) + IRF + 국면 forward
summary = dartlab.macro("종합", market="KR")
transmission = dartlab.macro("전파", market="KR", sectorKey="semiconductor")
c = dartlab.Company("005930")
company_macro = c.macro("사이클") # 시장 매크로, KR 자동
company_transmission = c.macro("전파") # Industry 위치와 Analysis 직접 근거 자동 바인딩
sensitivity = c.analysis("macro", "매크로민감도") # 기업 단위 매크로 민감도는 analysis 엔진
강행 호출 룰 (agent 답변 품질 회귀 차단)
15 axis 매크로 질문 (cycle·rates·transmission·corporate·trade 등) 에서 다음 4 룰 강행.
- 1 차 도구는 EngineCall 강제.
EngineCall(apiRef="macro", args={"axis": "rates", "market": "KR"})양식. RunPython 직접 ECOS/FRED 호출 금지 - 본 엔진이 HF SSOT 캐시 + tableRef 발급 담당. - 본문 숫자에
[valueRef:...]또는[dateRef:...]inline 표기 필수. macro 데이터는 시점 (asOf) 변동 큼 - dateRef 누락 시 stale 데이터 환각. - cycle / inventory 4 phase 판정은
[conf:30]기본 - 회고적 신호임을 명시. NBER vs ECRI vs Cleveland Fed 정의 차이 인지. - 단일 지표로 사이클 단정 금지 - CLI·LEI·yield curve 중 최소 2 종 ref 동행. 단일 지표 답변은 한계 명시 필수.
- macro EngineCall 결과는 본문에 최소 1 개 수치 + dateRef inline 인용 의무. macro 호출 했는데 답변에 결과 인용 0 회 = evidence flow 누락 회귀 (2026-05-20 OAuth probe 시나리오 F: rates/KR 3 회 호출 후 답변에 금리 수치 0 회 인용, NIM 단정 불가로 회피). 호출 결과의 핵심 지표 (예: KR base rate 3.25% [dateRef:date:macro:rates:KR:2026-Q1]) 1~2 개는 답변 본문 첫 단락에 inline. 결과 부족하면 그 이유 + 어떤 ref 필요 한계 명시.
산업별 macro 연결 - rates / liquidity / trade
기업 질문 + macro axis 결합 시 주로 등장하는 산업 매핑. 본 매핑은 직접 결합 규칙 아닌 대표 패턴 - 실제 결합은 c.analysis("macro","매크로민감도") + 시나리오 호출로.
| macro axis | 영향 큰 산업 | 결합 시 인용할 macro 지표 | 기업 측 인용할 재무 지표 |
|---|---|---|---|
| rates | 은행·보험·증권·리츠·고PER 성장주 | base rate / 10Y-3M / spread | 은행 NIM·예대마진, 증권 운용수익, 리츠 D/E |
| liquidity | 자산운용·증권·중소형 성장주 | M2 증가율 / NFCI / 신용스프레드 | 거래대금, 차입 규모, 변동성 |
| trade | 반도체·자동차·조선·정유 | 수출증가율 / 교역조건 / USD-KRW | 수출 비중·해외 매출, FX 손익 |
| crisis | 금융·부동산·고레버리지 | Credit-to-GDP gap / Minsky / GHS | D/E, 이자보상배율, 단기차입 |
| assets | 자산운용·증권·고배당 | 5 자산 배분 / Cu/Au / 위험선호 | ROE, 배당수익률, beta |
What ships with it
7 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 248 lines · 0 tokens per session scan A c2703adc1dd5
macro is a skill published in the GitHub repository eddmpython/dartlab (209 stars, last pushed 10d ago), licensed Apache-2.0. It costs nothing until one of its globs matches a file; then it loads 4,570 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
build_workspace_app
This guide covers the full lifecycle of building, running, and serving a custom OpenBB Workspace application from an extension project scaffolded by openbb-cookiecutter. It assumes the project shell already exists (see the developextension skill for scaffolding instructions).
configure_mcp_server
This guide covers installation, configuration, authentication, tool discovery, prompt management, and client integration for openbb-mcp-server.
develop_extension
This is a complete guide for creating a new OpenBB Platform extension from scratch. Follow every phase in order. When the user says "build me an application that does X", use this guide to scaffold, implement, install, and verify the extension.
work_with_server
This guide explains how to call tools, interpret responses, discover capabilities, use prompts, and handle errors when interacting with an OpenBB MCP server.
fin-data-acquisition
根据REFINEDDESIGN.md中的变量定义，自动获取所需数据并生成可执行的回归分析脚本（Python/Stata）。.
alphasift
自动选股 Skill。Use when: 用户要按策略筛选 A 股、列出可用策略、运行双低/放量突破/均衡多因子/资金热度等选股，或保存运行并做 T+N 后验评估。通过 alphasift CLI 或 Python 接口输出候选股票列表。.