A MCP (Model Context Protocol) server for agent-driven research on Brazilian law using official sources.
This server empowers models with scraping capacities, thus making research easier to anyone legitimately interested in Brazilian legal matters.
This facility comes with a price: the risk of overloading the official sources' servers if misused. Please be sure to keep the load on the sources to a reasonable amount.
- git
- uv (recommended) or Python >= 3.12
- Google Chrome
- Clone the repository:
git clone https://github.com/pdmtt/brlaw_mcp_server.git
- Install the dependencies
uv run patchright install
- Setup your MCP client (e.g. Claude Desktop):
{
"mcpServers": {
"brlaw_mcp_server": {
"command": "uv",
"args": [
"--directory",
"/<path>/brlaw_mcp_server",
"run",
"serve"
]
}
}
}
StjLegalPrecedentsRequest
: Research legal precedents made by the National High Court of Brazil (STJ) that meet the specified criteria.TstLegalPrecedentsRequest
: Research legal precedents made by the National High Labor Court of Brazil (TST) that meet the specified criteria.StfLegalPrecedentsRequest
: Research legal precedents made by the Supreme Court (STF) that meet the specified criteria.
The project uses:
- Ruff for linting and formatting.
- BasedPyright for type checking.
- Pytest for testing.
Resources, tools and prompts related stuff must be written in Portuguese, because this project aims to be used by non-dev folks, such as lawyers and law students.
Technical legal vocabulary is highly dependent on a country's legal tradition and translating it is no trivial task.
Development related stuff should stick to English as conventional, such as source code.
This project is licensed under the MIT License - see the LICENSE file for details.