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

Annotations declare idempotent, destructive and openWorld, and the description adds real context on top: the charge is drawn from the prepaid balance at the exact quoted total, and 'queued' explicitly means not printed, mailed or delivered yet. It also explains idempotent-retry mechanics (reuse key/body) and the INSUFFICIENT_FUNDS outcome. It does not say whether a queued order can still be cancelled or reversed, which matters for a destructiveHint tool.

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?

Four tight sentences, front-loaded with the core action, then the queued-vs-delivered caveat, then retry mechanics, then the error path. No filler and nothing repeated from the schema or annotations.

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 destructive, money-spending tool with no output schema, the description covers cost source, delivery semantics, idempotency and one error branch. It still leaves the required identifier chain (send_intent_id → quote_id → proof_id) and the successful response shape undocumented, which an agent needs to call this correctly on the first attempt.

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% for 7 parameters, so the description must carry the load. It clarifies grant_id (optional, permission fallback), expected_total_minor (must equal the exact final total) and idempotency_key (reuse on retry), but leaves send_intent_id, quote_id, proof_id and the USD const constraint unexplained and their relationships to this call unstated.

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?

Names a specific verb (send) and resource (the letter), then immediately disambiguates the verb's real semantics: it spends the exact final total from the prepaid balance and queues one mailing order. That distinguishes it cleanly from sibling steps quote_for_send, quote, prepare_checkout and create_payment_link, which an agent would otherwise confuse with 'send'.

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?

Gives clear operating context: grant_id is optional and falls back to the owner's active permission, and the same key/body must be reused on retry. It also prescribes an action on INSUFFICIENT_FUNDS (surface the returned funding link). What it lacks is an explicit statement of what must precede this call (the quote/proof flow) or which sibling to use instead when the user only wants pricing.

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