# Brazilian Judiciary Dictionary MCP > Brazilian Judiciary Dictionary Engine uses deterministic regex to find every Brazilian court, tribunal, agency, and regulatory body in any document. It checks against 100+ pre-indexed entities—STJ, ANVISA, BACEN, etc.—with zero AI guesswork or false positives. ## Overview - **Category:** knowledge-management - **Price:** Free - **Tags:** entity-extraction, regex, judiciary-data, data-normalization, legal-research, text-parsing ## Description Legal documents are a mess of acronyms and proper nouns; standard language models often fail when counting or identifying specific Brazilian courts. This MCP fixes that by running strict dictionary matching against the entire Brazilian legal structure. It identifies everything from Superior Tribunals (like STF) to state-level agencies, including regulators like CADE and financial bodies such as BACEN. You feed it a large batch of text, and it returns a count and list of every specific entity found, categorized by its type. Because this process uses pure dictionary matching, you get reliable results that don't rely on the AI guessing what an acronym means. Accessing this power through Vinkius allows your agent to bring structural accuracy to otherwise messy legal data. ## Tools ### search_legal_entities Scans text and finds known Brazilian legal entities (courts, tribunals) using a strict offline dictionary search. ## Prompt Examples **Prompt:** ``` Scan this 300-page litigation bundle and tell me exactly which Brazilian courts and agencies are referenced. ``` **Response:** ``` Scan complete. Found 9 entities: STJ (45 mentions), TJSP (23), TRF3 (18), CADE (12), MPF (8), CNJ (5), TRT2 (4), ANVISA (3), AGU (2). Category summary: {"Superior": 45, "TJ": 23, "TRF": 18, "Regulador": 15, "MP/Advocacia": 10, "Controle": 5, "TRT": 4}. ``` **Prompt:** ``` I need to know how many times each TRT appears in this labor law case file to determine jurisdictional concentration. ``` **Response:** ``` Analysis complete. TRT2 (São Paulo — Capital) leads with 34 mentions, TRT15 (Campinas) has 12, TRT1 (Rio de Janeiro) has 8, and TST has 19. Category breakdown: {"TRT": 54, "Superior": 19}. ``` **Prompt:** ``` Check if any regulatory agency (CADE, CVM, BACEN, ANVISA) is mentioned in this corporate compliance report. ``` **Response:** ``` Found 4 regulatory entities: BACEN (22 mentions), CVM (15), CADE (9), SUSEP (3). No mention of ANVISA. Category summary: {"Regulador": 49}. ``` ## Capabilities ### Identify all Brazilian judicial bodies It scans text and reports every instance of Superior Courts, Regional Tribunals (TRF), and State Courts (TJ) mentioned. ### Locate regulatory agencies The engine pulls out specific regulatory organizations, including the CNJ, CVM, ANVISA, and multiple other state/federal bodies. ### Count entity mentions by category It gives a summary count of how many times different types of entities—Superior, Regional, Regulatory—appear across the document set. ## Use Cases ### Reviewing Litigation Scope A paralegal needs to know exactly which courts are involved in a massive 300-page case file. Running the MCP quickly finds and counts all mentions of STJ, TJSP, and TRF3, giving them an immediate summary of jurisdictional involvement. ### Compliance Audit Check A compliance officer receives a corporate report and must confirm if key regulatory bodies like CADE or BACEN are mentioned. The MCP finds these specific agencies instantly, confirming coverage without manual searching. ### Labor Law Analysis A data analyst is studying labor law cases and needs to know which regional tribunals (TRT) dominate the discussion. They run the tool across all files, identifying and quantifying mentions of TRT2 versus TST. ### Regulatory Mapping An internal team must track every mention of various federal agencies—MPF, CNJ, or ANATEL—across a set of documents. The MCP groups these findings by category (Controle/Superior/Regulador), making the mapping clear. ## Benefits - Guaranteed Accuracy: It uses strict regex matching, meaning you don't have to worry about the AI misinterpreting acronyms or guessing names. The results are deterministic. - Comprehensive Coverage: It indexes over 100 specific Brazilian entities, covering not just federal courts (STF) but also regulators like ANVISA and financial groups like CVM. - Structured Tallying: Instead of a simple 'yes/no,' the MCP counts every mention and organizes them by category—Superior, Regional, Regulatory, etc.—giving you immediate data visualization. - Efficiency Boost: You stop manually cross-referencing large documents against internal lists. The engine processes the entire scope in one go using the `search_legal_entities` tool. - Deep Scope: It handles a vast array of Brazilian judicial bodies, including all 6 TRFs and nearly every state court (TJ), making it ideal for national legal review. ## How It Works The bottom line is you get an accurate count of specific legal bodies without any AI guesswork or false positives. 1. You pass the engine a large body of text (e.g., a litigation file or compliance report). 2. The MCP runs pure regex boundary matching against its internal dictionary of over 100 Brazilian legal entities. 3. It outputs a clean, structured list showing every unique entity found and how many times it was mentioned, grouped by category. ## Frequently Asked Questions **Does search_legal_entities only work for federal courts?** No, it covers the entire Brazilian judiciary system. It includes Superior Courts (STF, STJ) as well as all state-level tribunals (TJs), making its scope comprehensive. **Can search_legal_entities find regulatory agencies?** Yes. The engine is specifically programmed to identify major regulatory bodies like CADE, CVM, ANVISA, and BACEN, treating them as distinct legal entities. **What if I need to search for a country other than Brazil?** This MCP focuses exclusively on the Brazilian judiciary. If you need another jurisdiction, you'll have to use its custom dictionary parameter or an alternative tool set up for that region. **How does search_legal_entities handle acronyms?** It uses strict regex boundary matching, which is a deterministic process. This means it finds the exact phrase match and won't be confused by context or general AI inference. **Does using `search_legal_entities` require the input text to be in plain format?** Yes, you must feed it clean, extracted text data. The engine only processes strings and cannot read PDFs, images, or complex file formats directly. **What are the performance considerations when running `search_legal_entities` on massive documents?** There are no hard-coded rate limits for volume. Processing time depends purely on the document's length and the number of unique entities it contains. **How does `search_legal_entities` handle slight variations, like hyphens or punctuation?** It requires exact matches based on predefined word boundaries. If an entity name is hyphenated differently or has surrounding punctuation not in the dictionary, it will not detect it. **Is the data processed by `search_legal_entities` secure and private?** Yes, processing happens locally using pure regex matching. The engine does not transmit or store your source document content anywhere. **What exactly is pre-indexed?** The complete Brazilian judiciary: 5 Superior Courts, 6 TRFs, 24 TRTs, 27 TJs (including TJDFT), 3 military TJMs, oversight bodies (CNJ, CNMP, TCU), prosecution and advocacy (AGU, MPF, MPT, MPM, MPDFT, DPU, OAB), and 15 regulatory agencies (CADE, CVM, BACEN, INPI, INSS, SUSEP, ANATEL, ANVISA, ANS, ANAC, ANEEL, ANP, ANA, ANTT, ANTAQ). **Does it cover courts from other countries?** No. This engine covers exclusively the Brazilian legal system. To add courts from other countries, pass a custom JSON dictionary mapping acronyms to full names. Custom entries are tagged separately in the output. **How are results organized?** Each detected entity includes its acronym, full official name, category (Superior, TRF, TRT, TJ, TJM, Controle, MP/Advocacia, Regulador), and exact mention count. A category summary is also provided for quick analysis.