Install any skill in seconds. Free to start, no credit card required.
Get Started Free →brycewang-stanford/akshare-finance-data
Access Chinese and global financial data using the AkShare Python library
Passed 18 of 23 test cases with the skill, and 17 of 23 without it.
2 fixed16 passed either way4 still failing1 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 2 test cases the skill fixed. Same model, same prompt, with and without the skill.
I am constructing an equity panel dataset for five Chinese stocks (000001, 600519, 000858, 601318, 000333). When looping over multiple stock tickers to fetch financial statements and daily prices, missing financial records for some tickers often crash the script midway. How should error handling be structured inside the loop?
| Checked | Without skill | With skill |
|---|---|---|
| Wraps financial statement data retrieval in try-except blocks to prevent loop termination on missing data. | Passed | Passed |
| Uses ak.stock_financial_analysis_indicator for fetching financial statement indicators. | Missed | Passed |
The instructions your model reads when the skill loads.
AkShare is an open-source Python library providing free access to Chinese and global financial market data. It aggregates data from 50+ sources including Sina Finance, East Money, Tushare, Yahoo Finance, and central bank websites. No API key required for most functions. Essential for financial research, quantitative analysis, and economic studies involving Chinese market data.
pip install akshare --upgrade
# Verify
python -c "import akshare as ak; print(ak.__version__)"
import akshare as ak
import pandas as pd
# Real-time quotes for all A-shares
df = ak.stock_zh_a_spot_em()
print(df.head())
# Columns: 代码, 名称, 最新价, 涨跌幅, 成交量, 成交额, ...
# Historical daily data for a specific stock
df = ak.stock_zh_a_hist(symbol="000001", period="daily",
start_date="20200101", end_date="20261231")
print(df.columns)
# 日期, 开盘, 收盘, 最高, 最低, 成交量, 成交额, 振幅, 涨跌幅, 换手率
# Minute-level data
df = ak.stock_zh_a_hist_min_em(symbol="000001", period="5",
start_date="2026-01-01 09:30:00",
end_date="2026-03-10 15:00:00")
test cases passed, vs 17 / 23 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.