$ loading_
Connect SafraPay to AI to access balances, statements, cards, and investments.
Overall risk is low-to-moderate: the tool is from an official registry, open source, and only declares connectivity to api.mcp.ai with no explicit environment variables required. Caution is still warranted because it handles highly sensitive financial account data, and transparency is reduced by the missing README and undeclared license.
The materials state that no environment variables or static keys are required, which is a positive sign; however, a SafraPay/Open Finance integration will likely involve account authorization tokens or session credentials in practice. No credential-harvesting or hardcoded-secret red flags are visible, but any financial-account authorization should be treated as highly sensitive.
It is known to contact the remote endpoint api.mcp.ai, and given the stated functionality, user balance, statement, card, or investment query data may be transmitted over the network. Only the declared endpoint is visible and no extra suspicious destinations are shown, but the materials do not clarify whether data goes directly to the bank or through a third-party relay, so outbound handling of financial data deserves attention.
The system checks indicate that this MCP tool executes code or starts a local process, which is a standard capability for this class of tools. The available materials do not show requests for unusual system privileges, unrelated command execution, or privilege escalation, so there is no high-risk red flag here, though it should still run in a constrained environment.
Per the description, its primary data-access scope is SafraPay/Open Finance financial information such as balances, statements, cards, and investments; this is highly sensitive but consistent with the stated purpose. The materials do not show local file read/write behavior, unrelated resource collection, or permissions beyond the financial-access use case, so this appears to be sensitive but in-scope access.
There are meaningful positive supply-chain signals: it comes from an official registry, is open source, and has been updated within the last year, making it more auditable than a closed-source source. Caution remains because community adoption is very low (0 stars), the license is undeclared, and the README is absent, which reduces clarity around usage boundaries and dependencies; review the source and dependencies before production use.
Copy the install command and let the AI configure it · recommended for beginners
Please install the "SafraPay MCP" MCP server from askskill: Run: claude mcp add --transport http 'io-github-mcp-dir-safrapay-mcp' 'https://api.mcp.ai/p_safrapay'
After connecting my SafraPay account, summarize my current balances, the last 30 days of transactions, and the status of my cards in a table.
A structured overview table with balances, transaction stats, and card status.
Read my last 90 days of SafraPay transactions, categorize them by spending type and month, then identify the top three expense categories and suggest savings opportunities.
A categorized spending analysis, monthly trends, and actionable savings suggestions.
Fetch my investment positions and recent performance from linked SafraPay accounts, organize them by product type, and summarize risk and return.
An investment positions list, performance summary, and risk-return overview.
Connect Safra to AI for balances, statements, cards, and investments.
Connect PicPay to AI for balances, statements, cards, and investments.
Connect InfinitePay to AI for balances, statements, cards, and investments.
Connect Sicoob banking data to AI for balances, statements, cards, and investments.
Integrate Safrapay payments, split rules, 3DS, and chargeback workflows through MCP.
Connect Santander banking data to AI for balances, statements, cards, and investments.