prometeo

Prometeo MCP Server provides a Python-based MCP (Model Context Protocol) gateway that exposes the Prometeo API capabilities to MCP-enabled LLMs and tools. It implements functions for banking operations, CURP queries, account validation, and cross-border actions, enabling you to validate accounts, fetch movements, perform curp lookups, and interact with Prometeo’s banking and cross-border features through a standardized MCP interface. You can run the server in an MCP-enabled environment (such as Claude Desktop, LangChain tooling, or OpenInterpreter) and reference the available tools under the PrometeoAPI namespace to perform specific operations like curp_query, validate_account, banking_login, banking_get_accounts, crossborder_create_payout, and more. The server expects an API key and environment setting to authenticate with Prometeo’s services.

Prerequisites:

• Python 3.11+ installed
• uv (the UV runtime) installed globally
• Access to Prometeo API with a API key

Installation steps:

1. Clone the repository: git clone https://github.com/prometeoapi/prometeo-mcp.git cd prometeo-mcp

2. Set up a Python environment (recommended): uv venv source .venv/bin/activate

3. Install dependencies in editable mode (from repository root): uv pip install -e .

4. Run the MCP server via uv (requires API key): uv run prometeo_mcp/server.py

5. (Optional) If integrating with an MCP-compatible LLM, configure your host tool to reference the server using the recommended mcpServers entry (see mcp_config in this document).

Tips and considerations:

• Ensure PROMETEO_API_KEY is kept secret and provided in your environment configuration when launching the server.
• PROMETEO_ENVIRONMENT can be sandbox or production depending on your access; use sandbox for testing.
• The server exposes a suite of CURP, banking, and cross-border tools; refer to the Tools section of the README for exact function names and usage.
• When running inside LLM environments, you can specify the command and environment variables in the mcpServers configuration (as shown in mcp_config).
• If you encounter installation issues, verify that Python 3.11+ is being used and that uv is installed globally (uv must be able to execute run commands).
• Tests are run with pytest; ensure pytest and pytest-asyncio are installed if you plan to run tests locally.