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

Annotations already flag a non-read-only, non-idempotent, open-world operation, so the description carries the burden of richer context — and it delivers substantially: the MX-checked email/https webhook reply mechanism, the pollable-thread model (thread_url + access_token), and the security-critical disclosure that the token 'is shown only once.' This is exactly the kind of behavioral and operational context annotations cannot convey.

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?

Three dense sentences, each earning its place: purpose and scope first, then the reply_to requirement with its two delivery mechanisms, then the thread/token return behavior with the one-time-display warning. No padding, no filler, and the most operationally critical detail (token shown only once) is placed in the final sentence without being buried.

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?

With no output schema, the description carries the burden of explaining return values — and it does: the response carries thread_url + access_token, every message is a pollable thread, and check_message_thread is named as the follow-up path even without a mailbox. For a flat 4-parameter message-sending tool, nothing an agent needs to call it correctly and retrieve the reply 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 100%, so the baseline is 3. The description adds genuine meaning beyond the schema by specifying what belongs in message (questions, scoping, custom/recurring projects) and by explaining why reply_to is the delivery path for the reply. It does repeat some reply_to details that the schema already covers nearly verbatim, but the message-content guidance is a real addition above baseline.

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+resource pair ('Send a free-form message to the human operator') and immediately scopes it with 'anything that is not yet a ready-made task,' which cleanly separates it from siblings like submit_human_task. The free-form versus ready-made distinction is precise and actionable, so an agent knows exactly what this tool is for without opening other schemas.

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?

Provides explicit when-to-use context ('questions, scoping, custom or recurring projects') and names check_message_thread as the companion tool for reading replies. However, the primary alternative for ready-made tasks (submit_human_task) is never named; the exclusion is implied by 'not yet a ready-made task' rather than stated as an explicit when-not-to-use rule, which keeps it just shy of a 5.

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