Send email

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

Annotations only state non-read-only/non-idempotent/openWorld; the description adds far more: a human approval window with a ~50 second timeout, external-recipient flagging, decline/edit outcomes, the edited_by_user flag, the duplicate:true dedup behavior within 10 minutes, and the fact that a decline/timeout is surfaced as an error with nothing sent. This is exactly the behavioral context annotations cannot carry.

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?

Front-loaded with the core action and every sentence carries distinct operational information (approval, timeout, outcomes, dedup), so nothing is wasted. It is a dense single paragraph with no visual structure, which slightly hurts scannability for a definition this information-rich.

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 high-stakes outbound tool with an output schema already present, the description still covers the auth prerequisite (OWA Mail MCP setup allowance), the human-approval gate, timeout behavior, error semantics, and dedup. Nothing an agent needs to invoke it safely is missing.

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 coverage is 71% and the description only loosely adds meaning by noting what the confirmation window displays (recipients, subject, text). It does not clarify cc/bcc or the reply_to_message_id/reply_all interaction beyond what the schema already documents, so it is a baseline 3 with the schema doing most of the work.

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?

States a specific verb and resource (sends an email from the user's mailbox) and immediately establishes the scope that distinguishes it from every sibling, all of which are read-only lookup tools (get_message, search_messages, check_mail, etc.). An agent can tell this is the only mutating/outbound tool in the set without opening a schema.

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 explicit when-to-use ("Call only when the user asked to send this message") and explicit when-not-to-retry behavior for decline or timeout, plus the duplicate-call window. These are the exact conditions an agent needs to avoid unsolicited sends or retry loops.

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