Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare readOnlyHint=true, idempotentHint=true, and destructiveHint=false, and the description adds substantial context beyond them: snapshot freshness semantics, the meaning of updateDateTime/updatedAt, specific error codes (BALANCE_FETCH_ERROR, BALANCE_CONSENT_ERROR) with structured warnings, the consent-recovery path, and row-degradation behavior on upstream 5xx errors. This is exactly the kind of behavioral nuance an agent cannot infer from annotations alone.

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?

The description is long but every sentence carries distinct operational content: purpose, freshness semantics, remediation workflow, parameter bounds, credit-account error modes, and degradation behavior. It is front-loaded with the core purpose and organized so that an agent can scan to the relevant clause. No filler or redundancy.

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?

Given there is no output schema, the description does a good job of revealing partial return shape (rows contain updateDateTime/updatedAt, error rows carry a structured warning, degraded rows appear instead of throwing). The main remaining gap is the successful response payload for a normal row (exact balance field name and currency), which an agent would likely need to discover empirically. For a simple single-parameter read tool with rich error handling, this is nearly complete.

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?

Schema description coverage is 0%, so the description carries the full burden for the single parameter. It adds the critical constraint 'Pass account_ids as an array (1–50)', which is the most operationally important detail beyond the schema's bare 'array of strings'. However, it does not state where the account IDs come from (e.g., openfinance_list_accounts) or provide an example, so the guidance is strong but not exhaustive.

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 opens with a specific verb and resource: 'Returns the latest available balance per account id (GET /accounts/:id/balance).' It explicitly clarifies that this is a snapshot anchored to the last upstream sync, not a live read, which sharply distinguishes it from sibling tools like openfinance_list_transactions and openfinance_force_sync. An agent can immediately know what this tool does and what its key limitation is.

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?

The description gives explicit when-to-use and when-to-escalate guidance: if a recent movement is missing or the balance disagrees with the sum of openfinance_list_transactions, the agent is told to run openfinance_force_sync and then re-read. It also names the conditions under which the tool returns structured warnings instead of throwing, so the caller can handle CREDIT-account error cases appropriately.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.