Loading ToolBench
Ehliyet sinavi (Turkiye surucu belgesi) hazirlik verilerini AI ajanlarina sunan Model Context Protocol sunucusu. Provides Turkish driver's license exam preparation data including practice questions, quick facts, lesson summaries, and mock exams.
Server has 5 tools with complete input schemas and descriptions in Turkish. All tools are READ_ONLY and properly annotated. However, descriptions lack LLM-optimized structure (no WHAT/WHEN/WHY breakdown), parameter descriptions are minimal, and output schemas are not formally documented. Tool names follow verb_noun pattern but descriptions average ~180 chars, acceptable but could be more concise. No error handling guidance or recovery paths documented.
Belirtilen soru ID'si icin detayli aciklama dondurur. Dogru cevap, soru metni, secenekler, ilgili hap bilgiler ve ilgili dersler icerir.
50 soruluk tam bir deneme sinavi olusturur. Dagitim gercek MTSK sinaviyla aynidir: 12 Ilk Yardim, 23 Trafik ve Cevre, 9 Arac Teknigi, 6 Trafik Adabi. Gecme baraj 70 puan (35 dogru soru).
Belirtilen bolum ve/veya konu icin ders ozetlerini dondurur. Her ders ozeti: ders adi, bolum, URL ve temel bilgileri icerir.
Belirtilen bolumden rastgele ehliyet sinav sorulari dondurur. Her soru: soru metni, 4 secenegi, dogru cevabi ve soru kimligini icerir. Turkiye B sinifi surucu belgesi (ehliyet) sinavina hazirlik icin kullanilir.
Belirtilen bolum icin hap bilgileri (kisa ozetler) dondurur. Bunlar sinavda en cok sorulan bilgilerin ozetleridir. Hizli tekrar ve son dakika calismasi icin idealdir.
Output schemas not documented. Tool descriptions state what data is returned (e.g., 'soru metni, 4 secenegi, dogru cevabi') but no formal JSON Schema for responses. LLMs cannot plan downstream tool calls or extract fields reliably without structured output specs.
Parameter descriptions are minimal or absent. 'section' param has enum values but no explanation of what each section covers or when to use 'all'. 'count' param lacks guidance on typical values or performance implications. 'topic' param in get_lesson_summary is marked optional but no guidance on partial matching behavior.
No error handling or recovery guidance. If question_id is invalid in explain_answer, handler returns plain text error. No guidance on valid ID format, how to discover valid IDs, or retry strategy. Descriptions do not mention prerequisites (e.g., 'Call get_practice_questions first to get valid question IDs').
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 61 | 2026-07-28+ | v2 |
Tool descriptions lack LLM-optimized structure. Descriptions are in Turkish and state WHAT the tool does but omit WHEN to use it and WHY. No dependency hints (e.g., 'Use after get_practice_questions to explain a specific question'). Descriptions average 180 chars but lack actionable context.
generate_mock_exam has empty input schema (no properties). While this is valid for a no-argument tool, the description does not explain that no parameters are accepted or what the fixed distribution is. LLMs may attempt to pass parameters that are silently ignored.