Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/malda231125/TradingAgents-KRnpx agentmods add skills/malda231125/tradingagents-kr/tradingagents-korea-analysisWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/malda231125/tradingagents-kr/tradingagents-korea-analysis)<a href="https://agentmods.dev/skills/malda231125/tradingagents-kr/tradingagents-korea-analysis"><img src="https://agentmods.dev/badge/skills/malda231125/tradingagents-kr/tradingagents-korea-analysis/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/malda231125/tradingagents-kr/tradingagents-korea-analysis"><img src="https://agentmods.dev/badge/skills/malda231125/tradingagents-kr/tradingagents-korea-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What 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 5.1 | $0.00150 | $0.03749 |
| Opus 5 | $0.00075 | $0.01875 |
| Sonnet 5 | $0.00030 | $0.00750 |
| Haiku 4.5 | $0.00015 | $0.00375 |
Grade A, and why
tradingagents-korea-analysis 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 11d ago.
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 — 249 lines — stays where its author put it; the contents beside it link to each section on GitHub.
TradingAgents Korea Analysis
Use this skill when the user asks to analyze a Korean/KRX stock using the local malda231125/TradingAgents-KR repo and wants the long result as a Notion page.
Routing boundary
- Use this skill for Korean/KRX stocks only:
- 6-digit Korean stock codes such as
005930,000660 - Korean company names such as 삼성전자, SK하이닉스, 현대차
- requests mentioning 국내주식, 한국 주식, 코스피, 코스닥, KRX
- common phrasing:
- 삼성전자 국내 주식 분석해줘 → Korea
- 하이닉스 국내주식 분석해줘 → Korea
- 현대차 한국 주식 분석해줘 → Korea
- 6-digit Korean stock codes such as
- Do not use this skill for US/overseas stocks:
- tickers such as
AAPL,TSLA,NVDA,SPCX - requests mentioning 미국 주식, 해외 주식, NASDAQ, NYSE, US stock
- Korean names of overseas companies with 해외/미국 wording:
- 애플 해외 주식 분석해줘 → US
- 구글 해외 주식 분석해줘 → US
- 테슬라 미국 주식 분석해줘 → US
- use
tradingagents-us-analysisinstead
- tickers such as
- If the user only says “주식 분석” and the market is ambiguous, infer from ticker format first. Ask one concise market-clarifying question only if the ticker/name cannot be classified.
- See
references/routing.mdfor the Korea/US decision table.
What it does
- Resolves a Korean stock name/code from the request.
- Runs this repository’s self-loop workflow on the checked-out repo branch, normally
main. - Uses the repo
.venv,.env, and Codex CLI auth already configured in the OpenClaw container. - Must use the full TradingAgents workflow, not a one-shot replacement, unless the user explicitly asks for a quick packet-only draft.
- Must use a subscription CLI provider, not direct OpenAI API-key execution.
- Default production runs use Codex CLI. For Claude A/B or explicit user requests, use
--llm-provider claude-cli --quick-model sonnet --deep-model opus-4.8 --cli-effort max. - Before generating a new self-loop report, the wrapper automatically runs the daily Key Risk Trigger check against the same self-loop base directory. Do not create a separate scheduled trigger-check job unless the user explicitly asks for an independent audit run.
- Publishes the self-loop report to Notion, including the latest decision record, digest, and prior retrospective when available.
- Surfaces concrete up/down Key Risk Triggers and a
🎯 Key Risk Trigger 일치line when reporting trigger verification results. - Returns the Notion URL and a short summary.
- Can also read a prior Notion report URL and fact-check the report's predicted signals against later actual market data.
What ships with it
3 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.
- 11d ago First seen · 249 lines · 150 tokens per session scan A 0294ceae229d
tradingagents-korea-analysis is a skill published in the GitHub repository malda231125/TradingAgents-KR (6 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 150 tokens to every session and 3,749 once invoked, about $0.0007 per session on Opus 5. 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 2026-08-31.
Other skills, from other repositories
company
A single-company research interface for South Korean DART filings and United States EDGAR filings. It provides one object through which users can request company records, analysis, credit, prices, quantitative data, market context, industry information, and reports.
credit
A credit-risk analysis engine for one non-financial company. It assesses repayment, capital structure, liquidity, cash flow, business stability, financial reliability, and disclosure risk, then produces a dCR rating from dCR-AA+ to dCR-D.
panel
A data tool that turns government financial disclosures into a wide table, with financial items as rows and reporting periods as columns. DART is South Korea’s public company filing system, and XBRL is a standard format for labeling financial data.
quant
A quantitative analysis tool for prices, trading volume, market factors, risk, public filings, rankings, portfolios, and strategy tests. A backtest checks how a trading strategy would have performed on past data.
dataHub
A central catalogue and query layer for source data, cleaned data, cross-company data, and analysis products. It lets other programs discover and request these assets through a shared contract.
gather
An engine for collecting outside data used in financial analysis, such as prices, news, dividends, ownership, sector information, and economic indicators. It can gather data for Korean and US stocks and create snapshots of the source data.