What problem does it solve? Accessing Brazilian government, economic, judicial, electoral, and health data requires integrating dozens of fragmented public APIs with different formats and authentication schemes. This Skill consolidates 70+ Brazilian public data sources into a single MCP server so AI agents can query them through one unified interface. ## Core Features & Use Cases - 533 Tools Across 70 Data Sources: Query economy (Banco Central, IBGE), legislation (Câmara, Senado), transparency (Portal da Transparência, TCU), judiciary (DataJud), elections (TSE), environment (INPE, IBAMA), health (DataSUS, ANVISA), and education (INEP) data. - Smart Tool Discovery and Planning: BM25 search filters 533 tools down to relevant ones per query, and the planejar_consulta tool builds multi-API execution plans for cross-referencing data. - Local DuckDB Datasets: Cache large datasets like TSE candidate records, ANP fuel prices, and INEP school census locally for millisecond SQL queries. - Use Case: A journalist investigating a deputy can combine camara expense tools, transparencia contract queries, and TSE asset declarations in one conversation to cross-reference spending, voting records, and declared wealth. ## Quick Start Install the mcp-brasil package with pip or uv, then add the provided mcp-brasil server configuration to your Claude Desktop or MCP client config and ask the agent to query Brazilian economic indicators such as the Selic rate time series.
