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

Annotations already declare destructiveHint=true, openWorldHint=true and idempotentHint=false, but the description adds real context beyond them: policy gating, billing to the workspace balance, and the non-obvious permission requirement that returned body, metadata, error and idempotency_key need sms:read 'including on replays'. It stops short of saying whether a sent message can be recalled or what a policy denial returns, which keeps it out of 5 territory.

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?

A single dense paragraph that front-loads the action and the billing/scope facts, then moves to optional-field semantics and finally the permission caveat. Every sentence carries information, though the run-on structure makes the recording-versus-enforcing distinctions slower to parse than a short bulleted layout would.

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, open-world 12-parameter mutation with no output schema, the description covers billing, permission requirements and the recording semantics of the optional fields, which is more than most. It still omits failure modes (policy rejection, insufficient balance), whether an idempotency key can be supplied on input given that it is only mentioned as a return field, and any delivery-confirmation behavior.

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 only 25% across 12 parameters, so the description has to compensate and partly does: it clarifies that purpose, categories and user_authorization are optional and merely recorded, and that recipient_type is never enforced and is not even stored on the SMS row. However, body, to_phone_number, from_phone_number_id, project_id, campaign_id, relationship and user_requested remain undocumented in both the schema and the description, leaving a substantial gap.

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?

The opening clause names a specific verb and resource ('Send a policy-gated outbound SMS') and adds two scoping facts an agent needs: the message goes from one of the workspace's numbers and is charged to the workspace balance. This is enough to separate it from create_outbound_call and send_agent_template_message without opening either 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?

Usage is only implied: 'policy-gated' hints that evaluate_outbound_policy may need to run first, but the description never says when to call this instead of authorize_outbound or create_outbound_call, nor what to do when the policy check fails. An agent gets a general sense of the tool's role but no explicit routing rules.

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