Yargı MCP acts as a specialized bridge that allows AI assistants to navigate the complex world of Turkish law. Instead of users manually searching through various government portals for court rulings or legal precedents, this tool allows them to simply ask an LLM to find and analyze specific cases. It connects to a vast range of Turkish legal databases—including high courts like the Constitutional Court and specialized boards like the Data Protection Authority (KVKK)—and translates official documents into a format that AI can easily understand and summarize. For those needing more precision, the server provides deep filtering capabilities across 79 different judicial departments and boards. It utilizes high-performance optimizations, such as "Dual" and "Triple" API support, to ensure maximum coverage and reliability even if one source is unavailable. Developers can perform highly specific searches using ISO 8601 date ranges or exact phrase matching, while the server automatically converts complex legal PDFs and HTML files into clean Markdown. This ensures the data is immediately ready for analysis, comparison, or drafting legal documents within a chat interface. What makes this tool particularly valuable for developers is its extreme efficiency, featuring a 61.8% token reduction optimization that significantly lowers operational costs and speeds up AI response times. It is designed for seamless integration using the `uvx` package runner, making it compatible with various environments including Claude Desktop, 5ire, and Gemini CLI. By providing a standardized interface for 19 distinct legal tools, it removes the friction of building custom scrapers or handling multiple authentication layers, allowing developers to focus on building sophisticated legal-tech applications.
Category: Files, Documents & PDFs
Tags: court-rulings, kvkk, legal-tech, pdf-parsing, turkish-law
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex" * Mac/Linux (Terminal): curl -LsSf https://astral.sh/uv/install.sh | sh * Git: Required for Windows users. * Microsoft Visual C++ Redistributable: Required for certain Python packages on Windows. To run the server directly via uvx: bash uvx yargi-mcpclaude_desktop_config.json file: json { "mcpServers": { "Yargı MCP": { "command": "uvx", "args": [ "yargi-mcp" ] } } } 4. Restart Claude Desktop.~/.gemini/settings.json on Mac/Linux or %USERPROFILE%\.gemini\settings.json on Windows) and add the following: json { "mcpServers": { "yargi_mcp": { "command": "uvx", "args": [ "yargi-mcp" ] } } }yargimcp, Name to Yargı MCP, and Command to uvx yargi-mcp. 4. Save and enable the tool.search_bedesten_unified: Search across 5 court types with filters for 79 departments, dates, and exact phrases. * get_bedesten_document_markdown: Retrieve any document by ID in Markdown format. * UYAP Emsal Karar: * search_emsal_detailed_decisions: Search precedent decisions with detailed criteria. * get_emsal_document_markdown: Get specific precedent text. * Uyuşmazlık Mahkemesi (Court of Jurisdictional Disputes): * search_uyusmazlik_decisions: Form-based search for jurisdictional decisions. * get_uyusmazlik_document_markdown_from_url: Retrieve decision text via URL. * Anayasa Mahkemesi (Constitutional Court): * search_anayasa_unified: Unified search for Norm Review and Individual Applications. * get_anayasa_document_unified: Retrieve paginated decision content. * KİK (Public Procurement Board): * search_kik_decisions: Search procurement board decisions. * get_kik_document_markdown: Retrieve paginated decision text using Base64 IDs. * Rekabet Kurumu (Competition Authority): * search_rekabet_kurumu_decisions: Search for Merger, Acquisition, or Competition decisions. * get_rekabet_kurumu_document: Extract and convert PDF pages to Markdown. * Sayıştay (Court of Accounts): * search_sayistay_genel_kurul: Search General Assembly interpretative decisions. * search_sayistay_temyiz_kurulu: Search Appeals Board decisions. * search_sayistay_daire: Search department-level audit decisions.search_kik_decisions and get_kik_document_markdown tools, an analyst can quickly pull the full text of past board rulings based on specific keywords or company names involved in previous tenders. Example: A tender consultant asks: "Retrieve the KİK decisions from the past year regarding 'security service tenders' where the bid was rejected due to 'insufficient work experience documents' and explain the Board's reasoning."search_anayasa_unified tool allows researchers to browse massive volumes of AYM decisions and retrieve paginated Markdown content, which is ideal…Part of MCP Servers