$ loading_
Connect Safra to AI for balances, statements, cards, and investments.
This MCP tool is from the official registry and is open source, with no clear high-risk red flags identified. However, it targets banking/open-finance data and has network/code-execution capabilities, so it should still be treated as a sensitive financial integration.
The material states that no local API key or environment variable is required, which reduces static secret exposure. However, the tool connects to a Safra account and accesses open-finance data, which typically involves highly sensitive user banking authorization tokens or sessions.
It is known to access the remote endpoint api.mcp.ai. Based on the stated functionality, user financial data such as balances, statements, cards, and investments is likely transmitted over the network. The material does not explain the exact data flow, whether it connects directly to Safra/open-finance APIs, or the specific proxy role of api.mcp.ai, so outbound data boundaries should be reviewed carefully.
System checks indicate that the tool can execute code or start processes. This is a common MCP capability and by itself is not enough to rate it as high risk. The available material does not show requests for system permissions beyond its declared financial connection/query purpose, but it should still be run in a constrained environment.
Per the description, it can access highly sensitive financial data including balances, statements, cards, and investments. This access is consistent with its declared purpose. The material does not state any local file read/write scope or obvious over-collection, so this currently appears to be sensitive but purpose-aligned data access rather than clear over-privilege.
Positive indicators include an official registry source, an auditable open-source repository, and updates within the last year; these factors materially reduce supply-chain risk. Points to watch are the missing README, undeclared license, and 0 GitHub stars, which mean weaker public documentation and community validation, so source and dependency review is advisable before production use.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "Safra MCP" MCP server from askskill: Run: claude mcp add --transport http 'io-github-mcp-dir-safra-mcp' 'https://api.mcp.ai/p_safra'
After connecting my Safra account, summarize my current balances, income and expenses from the last 30 days, outstanding card charges, and investment holdings, then highlight key changes by category.
A structured financial overview with balances, transaction stats, card spending, and investment summaries.
Read this month’s Safra account transactions and card spending, categorize them into dining, transport, subscriptions, and shopping, then identify the top expenses and suggest savings opportunities.
A monthly spending analysis with category totals, unusual expenses, and actionable cost-saving suggestions.
Pull my current holdings, recent performance changes, and asset allocation from my Safra investment account, then create a concise summary suitable for a weekly report.
An investment performance summary covering holdings, return changes, and allocation overview.
Connect Sicredi to AI for balances, statements, cards, and investments.
Connect Banco do Brasil to AI for balances, statements, and card data.
Connect Sicoob banking data to AI for balances, statements, cards, and investments.
Connect Santander banking data to AI for balances, statements, cards, and investments.
Connect Caixa accounts so AI can access balances, statements, and account data.
Connect Banrisul to AI for balances, statements, cards, and investments.