Loading ToolBench
Data Commons MCP server for accessing statistical observations and searching indicators in the Data Commons knowledge graph.
Two tools with complete schemas and reasonable descriptions. Both tools have proper parameter typing and descriptions. However, descriptions are generic and lack actionable context for LLM selection. No error handling guidance, no output schema documentation, and missing parameter constraints (enums, ranges). Tool names follow verb_noun convention correctly. Schemas are present but lack depth, no pagination support despite search_indicators returning potentially large result sets.
Get observations for a variable and place from Data Commons. This tool retrieves statistical observations (data points) for a specific variable (statistical measure) and place (geographic location) from the Data Commons knowledge graph.
Search for data indicators and topics in Data Commons. This tool searches the Data Commons knowledge graph for statistical variables, topics, and other indicators that match your query.
Missing output schema documentation. Neither tool documents what fields the response contains or their types. LLMs cannot plan downstream calls or extract required data without knowing response structure.
search_indicators lacks pagination support. Tool accepts per_search_limit but no offset/cursor or total_count in response. Large result sets will blow context window without pagination.
No error handling guidance. Both tools raise ValueError for missing DC_API_KEY but provide no recovery hints. Descriptions lack WHEN to use each tool vs alternatives.
Parameter descriptions lack constraints. 'date' parameter accepts string but no format specified (ISO 8601? YYYY-MM-DD?). 'per_search_limit' has no min/max bounds. LLMs will guess invalid values.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 52 | <=2025-11-25 | v2 |
Descriptions are generic and lack actionable context. 'Get observations for a variable and place' does not explain WHEN to call this vs search_indicators, what prerequisites exist, or what the typical workflow is.