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

Annotations only flag the safety profile (write, openWorld, destructive, non-idempotent); the description adds substantial behavior beyond that: guardian-only consent for under-18s, AI-without-owner asks waiting quietly in Requests, and that a recipient can relabel a request_kind 'and that sticks'. It also explains that re-asking with request_id returns the same approval rather than duplicating it.

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-loads the core action in the first clause before the who/child/edge-case detail, and every sentence carries information. The opening sentence is long with several colon- and dash-separated lists, making it denser than ideal, but there is no filler.

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 an 11-param write tool with annotations but no output schema, the description covers the critical what/whom/edge-case behavior well (guardian routing, AI queuing, expiry-from-schema, request_id reuse). It does not describe what the caller receives back (approval id, status), which is the one meaningful gap given no output schema is provided.

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, but the description meaningfully enriches several params: the reachable set for 'to', the about+group pairing that routes a child's ask to the guardian, and the request_kind semantics (recipient rules decide outcome; relabeling persists). It adds cross-parameter meaning rather than restating field docs.

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 concrete verb (ask) and outcome (a signed yes/no, choice pick, or short answer recorded with who said what and when), plus who can be addressed (groups, Wayza adults, AI addresses, email). The 'signed record' artifact and multi-modal answer scope clearly separate it from plain-messaging siblings like send_message, so an agent can pick it without opening the 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 concrete usage contexts ('a consent slip, a sign-off, an agent's pause for a human') and a conditional rule for the child case ('give about and group: it goes to their parent or guardian, and only that person's own tap counts'). It never names alternatives (e.g., send_message for non-approval asks, or get_approval/reply_approval for follow-up), so the when-not side is left to inference.

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