Yargı MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_emsal_detailed_decisions | Use this when searching UYAP precedent decisions (Emsal). For lower court decisions and case law. |
| get_emsal_document_markdownA | Use this when retrieving full text of an Emsal precedent decision. Returns clean Markdown format. |
| search_uyusmazlik_decisionsA | Use this when searching jurisdictional dispute court (Uyuşmazlık Mahkemesi) decisions. Resolves conflicts between civil and administrative courts. |
| get_uyusmazlik_document_markdown_from_urlA | Use this when retrieving full text of an Uyuşmazlık Mahkemesi decision. Returns clean Markdown format. |
| search_anayasa_unifiedA | Use this when searching Turkish Constitutional Court decision records. Supports norm control decisions and individual application decisions. Norm control filters include reviewed norm metadata; results are court decisions. |
| get_anayasa_document_unified | Use this when retrieving full text of a Constitutional Court decision. Auto-detects decision type from URL. |
| search_kik_v2_decisionsA | Use this when searching Turkish public procurement disputes (KİK). Supports dispute, regulatory, and court decision types. |
| get_kik_v2_document_markdownA | Use this when retrieving full text of a KİK procurement decision. Returns document in Markdown format. |
| search_rekabet_kurumu_decisionsC | Use this when searching Turkish competition law and antitrust decisions (Rekabet Kurumu). |
| get_rekabet_kurumu_documentA | Use this when retrieving full text of a Competition Authority decision. Returns paginated Markdown format. |
| search_bedesten_unified | Use this for Turkish court decision records from Yargıtay, Danıştay, Local Courts, Appeals Courts, and KYB via Bedesten. Prefer narrow court_types over all courts. pageSize is intentionally fixed to 10 results per page. Bedesten is upstream rate-limited; avoid parallel repeated calls and wait retry_after seconds after 429 responses. |
| get_bedesten_document_markdownA | Use this when retrieving full text of a Bedesten search result by documentId. Counts against the same Bedesten upstream rate limit as search; after 429, wait retry_after seconds before retrying. |
| search_sayistay_unified | Use this when searching Turkish Court of Accounts (Sayıştay) audit decisions. Supports Genel Kurul, Temyiz Kurulu, and Daire decisions. |
| get_sayistay_document_unifiedA | Use this when retrieving full text of a Sayıştay audit decision. Returns clean Markdown format. |
| check_government_servers_health | Use this when checking if Turkish legal database servers are online and responding. |
| search_kvkk_decisions | Use this when searching Turkish data protection (KVKK/GDPR equivalent) decisions. For privacy, consent, and data breach cases. |
| get_kvkk_document_markdown | Use this when retrieving full text of a KVKK data protection decision. Returns paginated Markdown with metadata. |
| search_bddk_decisionsA | Use this when searching Turkish banking regulation (BDDK) decisions. For banking licenses, fintech, and payment services. |
| get_bddk_document_markdownA | Use this when retrieving full text of a BDDK banking regulation decision. Returns paginated Markdown format. |
| search_btk_decisions | Use this when searching BTK Board decisions (Bilgi Teknolojileri ve Iletisim Kurumu Kurul Kararlari). Supports decision title keywords, decision number, decision date, publication date, and related department filters. |
| get_btk_document_markdown | Use this when retrieving full text of a BTK Board decision PDF. Returns paginated Markdown. |
| search_gib_ozelge | Search Turkish GİB özelge records (Revenue Administration tax rulings) - 18k+ rulings on VAT, income tax, corporate tax, stamp duty interpretations. kanunNo filters the related law number; returned documents are özelge/tax ruling records. |
| get_gib_ozelge_document_markdownA | Retrieve full text of a GİB özelge (tax ruling) by numeric ID. Returns paginated Markdown (5000-char chunks) with title, reference number, date and law metadata. |
| search_sigorta_tahkim_decisionsA | Search Sigorta Tahkim Komisyonu (Insurance Arbitration Commission) decisions from Hakem Karar Dergisi journals (64 issues, 2010-2025). Covers insurance disputes: traffic, health, fire, DASK, life insurance. |
| get_sigorta_tahkim_document_markdownA | Retrieve full PDF content of a Sigorta Tahkim Komisyonu Hakem Karar Dergisi issue by number. Returns paginated Markdown. Issues 1-64 available (2010-2025). |
| search_within_sigorta_tahkim_issue | Search within a specific Sigorta Tahkim Komisyonu journal issue for keywords. Downloads the PDF, splits into individual decisions, and returns matching decisions with excerpts sorted by relevance. |
| searchA | Only for ChatGPT Deep Research. Searches Bedesten-supported Turkish court databases and returns OpenAI Deep Research compatible results (id, title, text, url). For regular MCP use, prefer search_bedesten_unified. This tool does not fetch document bodies during search so one query stays within Bedesten upstream rate limits; call fetch only for selected result IDs. |
| fetchA | Only for ChatGPT Deep Research. Retrieves one Turkish legal document by numeric Bedesten ID. For regular MCP use, prefer get_bedesten_document_markdown. This performs one Bedesten document request and avoids an extra metadata lookup to respect upstream rate limits. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 17 tools
Each tool targets a distinct legal domain (Emsal, Bedesten, Uyuşmazlık, Anayasa, KİK, Rekabet, Sayıştay, BDDK, GİB, Sigorta Tahkim) with clear search vs. get operations per domain. The two Deep Research tools (fetch, search) are explicitly labeled as for ChatGPT Deep Research, and their descriptions clarify they should not be used for regular MCP, eliminating ambiguity.
The naming follows a verb_noun pattern (search_* and get_*) but with inconsistencies: some get tools include 'markdown' while others do not (e.g., get_rekabet_kurumu_document vs. get_bddk_document_markdown), some have special suffixes like '_unified' or '_from_url', and two tools ('fetch', 'search') break the pattern entirely. The core pattern is recognizable but not uniform.
With 17 tools, the server is slightly above the ideal 3-15 range but still reasonable for a multi-jurisdiction legal research platform. Each tool serves a distinct domain, and the count is justified by the breadth of sources covered, though it may feel heavy for agents to navigate.
Several domains lack either a search or get tool: Emsal has only a getter, Sayıştay has only a getter, GİB has only a getter, and Anayasa has only a search (no document retrieval). Additionally, a referenced 'search_bedesten_unified' is not included, leaving gaps in the Bedesten workflow. The surface is incomplete for a comprehensive legal research tool.