Statistical Data Fetcher Skill

This skill allows you to retrieve grounded economic and social data, ensuring specific numbers in your reports are accurate and traced to an official source.

Capabilities

1. Search Indicators: Find the correct ID for a metric (e.g., "GDP per capita" -> NY.GDP.PCAP.CD).
2. Fetch Data: Get time-series data for specific countries.

Usage

Run the python script fetch_worldbank.py.

Arguments

• --search (optional): search term to find indicator IDs.
• --indicator (optional): The content indicator ID (e.g., NY.GDP.MKTP.CD).
• --country (optional): Country code (ISO 2-letter, e.g., US, CN, JP) or "all". Default "US".
• --date (optional): Date range (e.g., "2020:2025").

Example

# 1. Find the indicator for "population"
python3 fetch_worldbank.py --search "total population"

# 2. Get US and China population for last 5 years
# (Assuming found ID is SP.POP.TOTL)
python3 fetch_worldbank.py --indicator SP.POP.TOTL --country US;CN --date 2020:2025

Output Format

The script outputs a JSON object (or list):

[
  {
    "indicator": {"id": "SP.POP.TOTL", "value": "Population, total"},
    "country": {"id": "CN", "value": "China"},
    "countryiso3code": "CHN",
    "date": "2023",
    "value": 1411750000,
    ...
  }
]

Tips for the Agent

• Always Search First: Indicator IDs are specific. Don't guess GDP. Search for "GDP" and pick NY.GDP.MKTP.CD (current US$).
• Nulls: Real data often has gaps. Check for null values.