Server facts
faostat-mcp
Project
- Author
- berba-q
- Repository
- github.com/berba-q/faostat-mcp
- License
- MIT
- Category
- Data & Analytics
Tech
- Runtime / language
- python
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 18
- Last commit
- Data as of
Tags
- Python
- agriculture
- ai tools
- fao
- faostat
- food security
- llm
- mcp
- model context protocol
- un data
- undata
- unfao
Overview
Crop yields, food imports, nutrition indicators, and agricultural emissions are among the subjects available through faostat-mcp. It connects MCP-compatible assistants to FAOSTAT for researchers, journalists, policy analysts, and developers working with FAO statistics. A query can begin with discovering dataset groups and domains, inspecting their available dimensions, and finding country, commodity, or measure codes. Code searches flag ambiguous matches for user confirmation. Retrieval tools then support filtered statistics, country rankings, and report data, while a separate size estimate helps assess a query before fetching its results.
The tools distinguish element filter codes used for data queries from display codes used for rankings. Results can be returned as JSON objects, compact columnar data, or CSV, with field selection to narrow the response. Definitions, methodological metadata, related documents, and bulk-file listings provide context beyond individual values. Credential setup and token refresh support authentication, while caching, request throttling, and retries handle repeated requests and transient failures. The Python implementation communicates with compatible clients over stdio.
faostat-mcp setup notes
- faostat-mcp records stdio only; confirm a network transport upstream before planning remote access.