Loading ToolBench
MCP server for getting free stock images from multiple sources including Unsplash, Pexels, Pixabay, Freepik, Burst, and StockVault
Three tools with complete JSON schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (search_, get_, check_). Descriptions are present but generic (10-50 chars for some params). No parameter descriptions for enum values in sources array. Output schemas not documented, LLMs cannot predict response structure. Error handling minimal; no recovery guidance. Schemas are well-formed but lack depth in parameter constraints and descriptions.
Check the status and availability of stock image APIs
Get information about available free stock image sources
Search for free stock images from multiple sources
Output schemas not documented. LLMs cannot predict response structure or plan downstream tool calls. search_stock_images returns formatted text but structure is undocumented.
Parameter descriptions missing or trivial. 'sources' array has no description explaining what each enum value represents or when to use specific sources. 'limit' lacks guidance on performance impact.
No error handling guidance. API failures (missing keys, rate limits, timeouts) return None or generic errors. LLMs receive no recovery hints like 'Set UNSPLASH_ACCESS_KEY env var' or 'Try again in 60 seconds'.
Tool descriptions are generic and lack context. 'Search for free stock images from multiple sources' does not explain when to use this vs get_stock_image_sources, or what format results are returned in.
Inferred effective spec: <=2025-11-25.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 63 | <=2025-11-25 | v2 |
No pagination or result limits documented. search_stock_images caps at 20 per source but description does not state this. Large result sets could exhaust context window.