功能描述

Fetch professional stock market data from Yahoo Finance (yfinance) and SEC EDGAR. Use when: user asks about stock prices, market data, company financials, ea...

使用说明 (SKILL.md)

Finance Data Skill

Name: singa- finance
Author: 1shadow1

Fetch professional stock and financial data from Yahoo Finance (via yfinance) and SEC EDGAR (free public API).

Setup

Install the Python dependency (one-time):

pip install -r skills/finance-data/scripts/requirements.txt

SEC EDGAR requires a User-Agent header identifying the requester. Set the env var (optional — a default is provided):

export SEC_EDGAR_USER_AGENT="YourName [email protected]"

Ticker Formats

Market	Format	Example
US stocks	Symbol	`AAPL`, `MSFT`, `GOOGL`
Shanghai A-shares	Code.SS	`600519.SS` (Moutai)
Shenzhen A-shares	Code.SZ	`000858.SZ` (Wuliangye)
Hong Kong	Code.HK	`0700.HK` (Tencent)
Tokyo	Code.T	`7203.T` (Toyota)
London	Code.L	`HSBA.L` (HSBC)
ETFs / Indices	Symbol	`SPY`, `QQQ`, `^GSPC`, `^HSI`

Yahoo Finance Commands

All commands output JSON to stdout.

Current Quote

python3 scripts/yfinance_query.py quote AAPL

Returns: price, volume, market cap, P/E, EPS, 52-week range, moving averages, dividend yield, beta, profit margins, etc.

Historical Prices (OHLCV)

# Last month, daily
python3 scripts/yfinance_query.py history AAPL

# Last year, weekly
python3 scripts/yfinance_query.py history AAPL --period 1y --interval 1wk

# Last 5 days, 5-minute bars
python3 scripts/yfinance_query.py history AAPL --period 5d --interval 5m

Period options: 1d, 5d, 1mo, 3mo, 6mo, 1y, 2y, 5y, 10y, ytd, max Interval options: 1m, 2m, 5m, 15m, 30m, 60m, 90m, 1h, 1d, 5d, 1wk, 1mo, 3mo

Financial Statements

# All statements (income, balance sheet, cash flow) — annual
python3 scripts/yfinance_query.py financials AAPL

# Quarterly income statement only
python3 scripts/yfinance_query.py financials AAPL --statement income --quarterly

# Balance sheet only
python3 scripts/yfinance_query.py financials AAPL --statement balance

# Cash flow only
python3 scripts/yfinance_query.py financials AAPL --statement cashflow

Full Company Profile

python3 scripts/yfinance_query.py info AAPL

Returns all available metadata: sector, industry, description, officers, full-year financials, etc.

Shareholders

python3 scripts/yfinance_query.py holders AAPL

Returns institutional holders, major holders breakdown, and insider transactions.

Analyst Recommendations & Price Targets

python3 scripts/yfinance_query.py analysts AAPL

Returns recent analyst recommendations (buy/hold/sell) and consensus price targets.

Dividend History

python3 scripts/yfinance_query.py dividends AAPL

Options Chain

# Nearest expiry (default)
python3 scripts/yfinance_query.py options AAPL

# Specific expiry date
python3 scripts/yfinance_query.py options AAPL --expiry 2025-06-20

Returns calls and puts with strike, bid, ask, volume, open interest, implied volatility.

Earnings

python3 scripts/yfinance_query.py earnings AAPL

Returns quarterly and annual earnings (revenue, earnings, surprise).

News

python3 scripts/yfinance_query.py news AAPL

Returns recent news headlines with links.

SEC EDGAR Commands

Free public API — no API key required. All US public company filings.

Full-Text Search

# Search across all filings
python3 scripts/sec_edgar.py search --query "artificial intelligence" --limit 10

# Filter by form type and date
python3 scripts/sec_edgar.py search --query "revenue growth" --form-type 10-K --date-from 2024-01-01

Company Filings

# Recent filings by ticker
python3 scripts/sec_edgar.py filings AAPL --limit 20

# Filter by form type
python3 scripts/sec_edgar.py filings AAPL --form-type 10-K
python3 scripts/sec_edgar.py filings AAPL --form-type 10-Q
python3 scripts/sec_edgar.py filings AAPL --form-type 8-K

# By CIK number
python3 scripts/sec_edgar.py filings 320193 --form-type 10-K

Common form types: 10-K (annual), 10-Q (quarterly), 8-K (current events), DEF 14A (proxy), S-1 (IPO), 13F-HR (institutional holdings).

Read Filing Content

Download and extract readable text from an SEC filing. Use the url from the filings command output.

# Read full filing (first 50000 chars by default)
python3 scripts/sec_edgar.py read-filing --url "https://www.sec.gov/Archives/edgar/data/320193/000032019325000079/aapl-20250927.htm"

# Read a specific 10-K section (much more focused)
python3 scripts/sec_edgar.py read-filing --url "URL" --section 7    # MD&A
python3 scripts/sec_edgar.py read-filing --url "URL" --section 1    # Business
python3 scripts/sec_edgar.py read-filing --url "URL" --section 1a   # Risk Factors
python3 scripts/sec_edgar.py read-filing --url "URL" --section 8    # Financial Statements

# Control output length
python3 scripts/sec_edgar.py read-filing --url "URL" --max-chars 100000

10-K section numbers: 1 (Business), 1a (Risk Factors), 2 (Properties), 5 (Market), 7 (MD&A), 7a (Quantitative Disclosures), 8 (Financial Statements), 9a (Controls), 10 (Directors), 11 (Executive Compensation), 12 (Security Ownership).

Recommended workflow for reading annual reports:

# Step 1: find the latest 10-K filing
python3 scripts/sec_edgar.py filings AAPL --form-type 10-K --limit 1

# Step 2: read a specific section using the url from step 1
python3 scripts/sec_edgar.py read-filing --url "\x3Curl_from_step_1>" --section 7

Filing Document Index

List all documents within a filing package (useful for finding exhibits, XBRL files, etc.):

python3 scripts/sec_edgar.py filing-index AAPL --accession "0000320193-25-000079"

Company Metadata

python3 scripts/sec_edgar.py submissions AAPL

Returns company name, CIK, SIC code, address, phone, fiscal year end, exchanges, and filing count.

Structured Financial Data (XBRL)

# List all available XBRL concepts for a company
python3 scripts/sec_edgar.py company AAPL

# Get time-series for a specific concept
python3 scripts/sec_edgar.py concept AAPL --concept Revenue
python3 scripts/sec_edgar.py concept AAPL --concept NetIncomeLoss
python3 scripts/sec_edgar.py concept AAPL --concept EarningsPerShareBasic
python3 scripts/sec_edgar.py concept AAPL --concept Assets
python3 scripts/sec_edgar.py concept AAPL --concept StockholdersEquity

Useful XBRL concepts: Revenue, NetIncomeLoss, EarningsPerShareBasic, EarningsPerShareDiluted, Assets, Liabilities, StockholdersEquity, OperatingIncomeLoss, CashAndCashEquivalentsAtCarryingValue, LongTermDebt.

Insider Transactions

python3 scripts/sec_edgar.py insider AAPL --limit 20

Returns Forms 3, 4, 5 filings (insider buys/sells/grants) with links to SEC documents.

Common Workflows

Quick stock check:

python3 scripts/yfinance_query.py quote TSLA

Fundamental analysis:

python3 scripts/yfinance_query.py financials MSFT --statement income
python3 scripts/yfinance_query.py financials MSFT --statement balance
python3 scripts/sec_edgar.py concept MSFT --concept Revenue
python3 scripts/sec_edgar.py concept MSFT --concept NetIncomeLoss

Read the latest annual report (10-K):

# 1. Find the latest 10-K
python3 scripts/sec_edgar.py filings AAPL --form-type 10-K --limit 1
# 2. Read the MD&A section (most important for investors)
python3 scripts/sec_edgar.py read-filing --url "\x3Curl>" --section 7
# 3. Read the risk factors
python3 scripts/sec_edgar.py read-filing --url "\x3Curl>" --section 1a

Due diligence / SEC filings:

python3 scripts/sec_edgar.py filings NVDA --form-type 10-K --limit 5
python3 scripts/sec_edgar.py insider NVDA --limit 20
python3 scripts/sec_edgar.py submissions NVDA

Compare multiple stocks:

Run quote or financials commands for each ticker and compare the JSON output side by side.

Notes

yfinance data is sourced from Yahoo Finance — may have a 15-minute delay for real-time prices
SEC EDGAR is official SEC data — always accurate but filings may lag by a few days
SEC rate limit: max 10 requests/second per IP; the scripts respect this by default
Chinese A-share tickers use .SS (Shanghai) or .SZ (Shenzhen) suffix in yfinance
XBRL concepts are case-sensitive (e.g. Revenue not revenue)

安全使用建议

This skill appears to do what it says: it runs the included Python scripts to fetch Yahoo Finance and SEC EDGAR data and only needs python3 and the yfinance package. Before installing: (1) review and run the scripts in a safe environment if you can, (2) set SEC_EDGAR_USER_AGENT as recommended to comply with SEC policy, (3) be cautious when using the read-filing command — it will fetch the URL you provide, so don’t pass internal or sensitive network addresses, and (4) keep the yfinance dependency up-to-date. There are no unexpected credential requests or install steps.

能力评估

✓ Purpose & Capability

Name/description match the included scripts and examples: the skill only needs python3 and yfinance to query Yahoo Finance and uses urllib to query SEC endpoints. No unrelated credentials, binaries, or config paths are requested.

ℹ Instruction Scope

Runtime instructions and CLI examples are focused on finance data. The SEC component accepts a URL for read-filing (intended to be the SEC URL from filings output) — that allows fetching arbitrary URLs if supplied, which is expected for 'read-filing' but could be misused to fetch internal resources (SSRF-like risk) if callers provide arbitrary URLs.

✓ Install Mechanism

No packaged install; scripts are provided and SKILL.md instructs pip install -r requirements.txt. requirements.txt contains only yfinance, which is proportionate.

✓ Credentials

Only optional SEC_EDGAR_USER_AGENT is suggested (per SEC policy). No secrets, API keys, or unrelated environment variables are requested.

✓ Persistence & Privilege

Skill is not always-enabled and uses normal agent invocation. It does not request persistent system-wide changes or other skills' credentials.

版本历史

v1.0.0

finance-data 1.0.0 - Initial release. - Fetches professional stock and financial data from Yahoo Finance (yfinance) and SEC EDGAR. - Supports real-time quotes, historical prices, full company profiles, financial statements, holders, analyst recommendations, dividends, options chains, earnings, and news. - Provides comprehensive SEC filings search, full-text filing reading with section extraction, filing document index, and company metadata. - Enables access to structured financial data (XBRL concepts) and insider transactions. - Compatible with US stocks, Chinese A-shares, and major international markets.

元数据

Slug finance-data

版本 1.0.0

许可证 MIT-0

累计安装 2

当前安装数 2

历史版本数 1

常见问题

singa- finance 是什么？

Fetch professional stock market data from Yahoo Finance (yfinance) and SEC EDGAR. Use when: user asks about stock prices, market data, company financials, ea... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 173 次。

如何安装 singa- finance？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install finance-data」即可一键安装，无需额外配置。

singa- finance 是免费的吗？

是的，singa- finance 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

singa- finance 支持哪些平台？

singa- finance 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 singa- finance？

由 1shadow1（@1shadow1）开发并维护，当前版本 v1.0.0。

singa- finance