MCP.AI for IDE agents (Cursor, etc.): log in in the browser, copy the access token. Best: add it to this server's config as a header Authorization: Bearer <token> for a permanent, non-expiring connection. Or paste it here for a session-only login: call with { token: "" } after the user pastes, or with no args to get the link.
| Name | Required | Description | Default |
|---|---|---|---|
| token | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Discloses key behaviors: permanent vs session-only authentication, and that calling without args returns a link. Annotations already indicate idempotency, but description adds context on token handling.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences covering both methods. Dense but not overly long. Minor improvement could be clearer separation of the two use cases.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple authentication tool, the description covers login process, token usage, and session persistence. No output schema needed as it's a setup action.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The single 'token' parameter is fully explained: its content (JWT) and the two invocation patterns (with/without token). This compensates for the 0% schema coverage.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: authentication via browser login and token management. It distinguishes from sibling tools like 'connect' by focusing on credential provisioning.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
Explicitly describes two usage modes (permanent config vs session token) and when to call with or without arguments. However, lacks explicit alternatives or when-not-to-use scenarios.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Most openfinance_* tools target distinct resource/action pairs, but there is real overlap among list_accounts, get_accounts_detail, and get_account_balance, as well as between list_transactions and list_transactions_by_item, and between get_credit_card_bill and list_credit_card_bills. The descriptions are detailed enough to usually resolve ambiguity, but the number of near-overlapping read/status tools makes mis-selection plausible.
The 21 openfinance_* tools follow a clear openfinance_<verb>_<noun> convention with predictable verbs like list, get, update, force, disconnect, and search. The non-prefixed platform tools (authenticate, connect, marketlace, report_bug, show_version, toolkit_info) break the pattern but are few enough to keep the naming mostly coherent.
25 tools is at the heavy end for the stated banking domain, and several could reasonably be merged: get_account_balance/get_accounts_detail/list_accounts and list_transactions/list_transactions_by_item overlap in purpose. The breadth of resources covered accounts, cards, loans, investments, connections, and provider health helps justify the count, but it still feels slightly bloated.
The Open Finance workflow is thoroughly covered: connect discover, force sync, status checks, accounts, balances, transactions, credit-card bills, loans, investments, category correction, disconnection, and provider health. Edge cases like stale data, consent errors, provider incidents, and cross-connector asymmetries are explicitly handled, leaving no obvious dead ends for an agent.