fred-navigator
by kiszly
Search & ResearchBrief
Navigate FRED categories and series using fredapi, supporting natural-language queries with intent recognition.
Description
What This Skill Does
FRED Navigator finds and retrieves economic data from the Federal Reserve Economic Data (FRED) database. It accepts a category ID, series ID, or natural-language query and returns matching series with metadata like frequency, units, and latest values. Natural-language queries go through intent recognition and two-stage validation before results are returned.
The two-stage validation (structural existence check plus semantic relevance scoring) reduces false matches when natural-language queries are ambiguous, which plain FRED search does not do.
When to Use It
- Looking up the current federal funds rate series by name
- Browsing all series under a specific FRED category by ID
- Searching for unemployment data using plain English
- Retrieving metadata and recent values for a known series like CPIAUCSL
- Exploring child categories within a FRED topic hierarchy
Example Workflow
User asks: find me inflation expectations data
- 1
Interprets query intent and selects best-matching FRED category candidate
- 2
Checks candidate exists in fred_categories_tree.json and has at least one series via scripts/fred_query.py check-category
- 3
Runs semantic comparison between query and candidate name/path, scores relevance
- 4
Accepts category if both checks pass, otherwise returns top-5 candidates for user selection
- 5
Retrieves series list via scripts/fred_query.py category <id> and returns id, title, frequency, units, and latest value
List of inflation expectations series with metadata including frequency, units, seasonal adjustment, and last updated date
Requirements
- FRED_API_KEY environment variable
Installation
clawhub install kiszly/fred-navigatorRun the CLI command, or paste the prompt into Claude Code, Cursor, Codex or whichever coding agent you use.