USDA Foreign Agricultural Service (trade & global production data)
FAS MCP provides 4 well-named, read-only agricultural data tools with complete input schemas and reasonable descriptions. All tools follow verb_noun naming (fas_exports, fas_imports, fas_production, fas_commodity_codes). Schemas include type definitions and required fields. However, output schemas are entirely undocumented, LLMs cannot predict response structure or plan downstream calls. Parameter descriptions lack format constraints (e.g., year format, country code format). Error handling is present but generic. No tool annotations (readOnlyHint, etc.). Missing pagination guidance for potentially large result sets.
List available USDA FAS PSD commodity codes with names and categories. Use these codes with fas_production and other FAS tools. Supports filtering by category or keyword.
Get US agricultural export data by commodity and destination country. Uses USDA FAS GATS (Global Agricultural Trade System) data. Shows export volumes and values.
Get US agricultural import data by commodity and origin country. Shows import volumes and values from USDA FAS trade data.
Get world production, supply, and distribution estimates for agricultural commodities from USDA FAS PSD (Production, Supply & Distribution) database. Covers global production, consumption, stocks, and trade flows.
Output schemas completely undocumented. LLMs cannot predict response structure, field names, or types. Breaks tool chaining and forces agents to guess at response format.
Parameter descriptions lack format constraints. 'start_year' and 'end_year' should specify format (e.g., 'YYYY format, e.g., 2024'). 'country' should document code format (ISO 3166-1 alpha-2). Ambiguity invites invalid LLM input.
No pagination guidance. Tools query external APIs that may return large result sets. No mention of limits, pagination parameters, or result caps. Risks context window exhaustion.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
No tool annotations. Tools are read-only but lack readOnlyHint annotation. Agents cannot distinguish safe tools from destructive ones without explicit metadata.
Error messages reference alternative data sources (NASS, EIA, FRED) but do not provide tool names or clear recovery paths. 'Use NASS Quick Stats (nass_query)' assumes nass_query exists in the agent's toolkit.