Fetch economic data from FRED, World Bank, BLS, OECD, and Yahoo Finance
SkillScore83Maintained
Automatic
◆Data Engineer pack
Last verified 8/7/2026
What it changes
Passed 19 of 23 test cases with the skill, and 15 of 23 without it.
Without skill65%
With skill83%
Checks it passed that the model alone missed
Uses wbdata to query World Bank indicator data.
Imports and calls imf_reader to retrieve IMF series.
Passes a dictionary mapping indicator codes to clean column names into wbdata.
and 2 more in the full report
5 fixed14 passed either way3 still failing1 broke
Pass-rate lift
+17 pp
Tokens
+241%
Agent turns
0%
Response time
-17%
Model gemini-3.6-flashMeasured 8/7/2026v119 of 23 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 5
Prompt
Write a Python script to pull national accounts expenditure GDP data for USA and GBR from the OECD REST API for 2010 through 2023 using requests and pandas.
What was checked, without and with the skill
Checked
Without skill
With skill
Queries the sdmx.oecd.org/public/rest/ endpoint instead of the deprecated stats.oecd.org endpoint.
Missed
Passed
case-04 · gemini-3.6-flash · 8/7/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md780 linessingle file
Data-Fetcher
Purpose
This skill helps economists fetch data from major economic data APIs including FRED (Federal Reserve Economic Data), World Bank, BLS (Bureau of Labor Statistics), OECD, and Yahoo Finance. It generates clean, documented Python code with proper error handling.
When to Use
Downloading macroeconomic indicators
Building custom datasets from multiple sources
Automating data updates for ongoing projects
Fetching cross-country panel data
Instructions
Step 0: API Key Setup Check (Run Before Anything Else)
Before generating any code, Claude must check for required API keys.
Read the file [plugin_root]/.env (same directory as .mcp.json).
Wait for input, then append FRED_API_KEY=<value> to .env.
If BLS_API_KEY is missing:
Inform the user it's optional but increases BLS rate limits, and they can get one free at https://www.bls.gov/developers/. If they want to add it later, just paste it and say "save my BLS key".
If .env exists and keys are already set: load them silently and inject them into all generated code via python-dotenv. Use load_dotenv() with no arguments so Python searches up from the current working directory automatically — never hardcode the plugin root path:
from dotenv import load_dotenv
load_dotenv() # searches CWD and parent directories for .env
The .env file stores keys locally and is never committed to version control. Generated scripts always read keys from environment variables — never hardcoded.