Submit a task to the human

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

Annotations (readOnlyHint=false, openWorldHint=true) already signal a real-world write, and the description adds substantial behavior beyond them: the human reviews every task before accepting, this is not instant execution, the operator is push-notified on submission, MX-checking on contact_email, and the pilot is free. This is exactly the kind of async and validation context an agent needs and that structured 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 with zero waste, and the most critical caveat (human review, not instant execution) is front-loaded. Every clause earns its place: async review, push notification, pilot free, MX-check, and the status_poll fallback all carry distinct information.

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 complex open-world tool (9 params, no output schema), the description is remarkably complete: it covers the return value (task_id), the async acceptance flow, the tracking path via check_task_status, both delivery modes with their constraints, and the daily budget. The only minor omission is what happens on task rejection, which is acceptable for a submission tool.

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% and the schema descriptions are themselves rich (delivery enum, task_type guidance). The description adds genuine cross-parameter value on top: the contact_email↔delivery coupling (email needs a real MX-checked mailbox; status_poll is the no-mailbox path), the 1-per-day status_poll budget, and 'keep the task_id, it is your only key.' That goes beyond a baseline 3 but the schema already does heavy lifting, so 4 is right.

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 ('Submit a task for the human operator to perform in the real world') and immediately scopes it as an async submission, not execution. It is clearly distinguishable from siblings: get_human_services (listing), check_task_status (tracking), message_human_operator (conversation). The core action, the return value, and the non-instant nature are all in the first sentence.

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?

Routes the agent to the correct follow-up tool ('check_task_status shows seen_by_operator_at...') and to get_human_services for task_type descriptions, and gives an explicit alternative for the delivery path ('No mailbox? Set delivery to status_poll instead'). It lacks explicit when-not guidance versus the messaging siblings (message_human_operator), so it earns a 4 rather than a 5, but the workflow routing is solid.

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