- DISCOVERY
- FILTERING
- REFINEMENT
- HUMAN
Short answer
How can I find specific FRED series IDs using AI?
Tell your AI what economic data you need in plain English. It uses metadata tags to filter the FRED database and returns the exact series IDs you need. You'll get the specific identifiers required for your analysis.
The results you get
Where your data discovery lands.
The AI handles the search logic so you can focus on the data itself.
DISCOVERY
Exact series IDs
The AI identifies the specific identifiers for the economic data you described. You get the IDs needed for direct API calls or spreadsheet imports.
FILTERING
Metadata-driven results
Instead of manual browsing, your agent uses specific tags to narrow down the massive FRED dataset. This eliminates the noise from irrelevant series.
REFINEMENT
Targeted source searches
You can isolate data coming from specific sources or categories. The AI maps your request to the correct metadata fields automatically.
HUMAN
Final data validation
The AI provides the IDs, but you decide if the series matches your research needs. You remain in control of which data enters your workflow.
The workflow
What your AI does when you ask for data.
The AI acts as a bridge between your questions and the FRED metadata structure.
Scan metadata
Your agent looks through available tags to see what categories of data exist in the FRED ecosystem.
search_tagsFilter by tags
Once it understands your request, the AI narrows the search to specific metadata attributes.
get_series_by_tagsIdentify IDs
The AI pulls the specific series IDs that match your filtered criteria.
get_series_by_tagsPresent results
The agent gives you a clean list of IDs and descriptions for your review.
search_tags
Try it
Copy these to start.
Use these prompts to begin your search.
Starting points
Swap out the bracketed terms for the specific economic indicators or sources you are looking for.
FRED Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_gSpxwBSlmBoTh0qkNEnTwJGqQpGQLWCmnpo1UhkQ/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — FRED capabilities are ready to use.
{
"mcpServers": {
"fred-tags-sources-data-discovery-mcp": {
"url": "https://edge.vinkius.com/vk_preview_gSpxwBSlmBoTh0qkNEnTwJGqQpGQLWCmnpo1UhkQ/mcp"
}
}
}Find all series IDs tagged with 'unemployment' and 'monthly'.
Search for economic series related to 'consumer price index' using available tags.
List the series IDs for data provided by the 'Federal Reserve Bank of St. Louis' source.
I need series IDs for inflation data; search through the tags to find them.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect FRED Tags & Sources once, then ask.
Click the link to connect the MCP to your client. Your credentials stay encrypted, and you can start querying FRED immediately.
Connect FRED Tags & Sources to your AIFAQ
How this task behaves
- 01
Can the AI download the actual data files?
No. This MCP is designed to find the correct series IDs using metadata. You use those IDs to pull the data through your preferred data tool.
- 02
Can the AI change or edit FRED data?
No. The AI only reads metadata and series information to help you find what you need. It has no write access to FRED.
- 03
Does it work with any economic database?
No. This specific MCP is built to interface with FRED metadata and tags.
- 04
Can I ask for data by name instead of tags?
Yes, but the AI uses the underlying tags to perform the search. You can just describe what you want in plain English.
- 05
Will the AI decide which data is best for my model?
No. The AI finds the IDs that match your description, but you are responsible for verifying that the data is appropriate for your specific research.
More questions about FRED Tags & Sources? See everything the FRED Tags & Sources Connector can do
Connect FRED to Claude, Cursor, ChatGPT & more
