Banco BRB MCP

The materials indicate no separate API key or environment variables are required, which is a positive sign. However, because it connects to Banco BRB/Open Finance accounts, real use will likely still involve user authorization tokens, sessions, or banking consent artifacts; these would be sensitive if exposed, and the documentation does not describe how they are stored or protected.

The tool is known to access the remote endpoint api.mcp.ai. Given its stated banking features for balances, statements, cards, and investments, user queries and related financial data are likely transmitted through that endpoint. No unrelated or suspicious egress target is shown in the materials, but the data involved is sensitive and should only be used if you trust this service path.

System checks show that it executes code or runs a local process, which is a normal property of MCP tools. The available materials do not indicate requests for system privileges beyond its stated function, nor obvious signs of malicious execution, but it should still be treated as having a local execution surface and deployed with least privilege.

Per the description, the intended data scope includes balances, statements/transaction history, cards, and investment information, all of which are highly sensitive personal financial data. These accesses are broadly aligned with the stated function, but the materials do not explain whether it reads or writes local files, caches responses, or retains history, so data minimization and retention behavior should be reviewed.

It comes from an official registry, has an open-source repository, and has been updated within the last year, all of which are positive factors that lower risk. However, the lack of a README, no declared license, and very low community adoption (0 stars) mean auditability and external validation are still limited, so supply-chain confidence is not strong enough for a fully low-risk rating.