Loading ToolBench
MCP server providing tools for Turkish traditional food recommendations, dish explanations, and menu planning with cultural context
Three tools with complete input schemas and descriptions. Tool names follow verb_noun pattern (recommend_, explain_, build_). Descriptions are adequate (100-150 chars) but lack actionable guidance on when to use each tool vs. alternatives. Parameters are well-typed with descriptions, but lack enums for constrained values (region, occasion, dietaryPreference accept free-form strings). Output schemas are not documented in the source code provided. No error handling guidance visible. Tool composition is sound, each does one thing, but parameter interdependencies (e.g., maxCookingMinutes + includeDessert) are not explicitly documented.
Build a small Turkish menu plan with starter, main and optional dessert using curated traditional food knowledge.
Explain one traditional Turkish dish with cultural context, category, dietary tags, estimated cooking time and common pairings.
Return a short, curated set of traditional Turkish food suggestions based on region, occasion, diet and cooking time. Do not return the full catalog.
No output schema documentation. Tool descriptions state what is returned (e.g., 'curated set of suggestions', 'cultural context') but the actual response structure (fields, types, pagination) is not visible in source code. LLMs cannot plan downstream operations without knowing response shape.
Free-form string parameters invite hallucinated values. 'region' accepts 'Aegean/Ege, Black Sea/Karadeniz, Central Anatolia/İç Anadolu, Gaziantep, Erzurum, Kayseri' as examples in description, but no enum constraint. LLMs will invent region names not in the knowledge base.
Parameter interdependencies undocumented. 'maxCookingMinutes' is optional but its interaction with 'occasion' (e.g., 'quick' implies time constraint) is not stated. 'includeDessert' in recommend_traditional_turkish_foods and build_turkish_menu_plan, unclear if both tools support it or if one is preferred.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 58 | 2026-07-28+ | v2 |
No error handling guidance. Tool descriptions do not explain failure modes (e.g., 'dishName not found', 'invalid region') or recovery steps. Agents cannot self-correct on invalid input.
Tool selection guidance missing. All three tools operate on Turkish food knowledge. Descriptions do not explain when to call recommend_traditional_turkish_foods vs. build_turkish_menu_plan (both accept region/occasion/dietaryPreference). LLMs may conflate them.