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

Annotations cover the safety profile (readOnlyHint=false, destructiveHint=false, openWorldHint=false), so the bar is lower, yet the description still adds real behavioural context: output is grounded only in facts in the brief, the result is a non-sending draft, and AI must be configured first. It stops short of saying what a repeated call produces (new draft each time, consistent with idempotentHint=false) or whether the draft is retrievable afterwards.

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 short sentences, front-loaded with what happens, immediately followed by the two constraints that matter most (nothing sent; AI prerequisite). Zero padding or restated boilerplate.

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 7-parameter tool with a deep nested audience object and no output schema, the description covers the essential mental model: AI writes from a brief, result is an editable draft, nothing is delivered. What is missing is any hint of the return value (e.g. draft id) and the audience-targeting dimension, which the schema carries alone.

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 86%, so the schema already documents brief, ctaUrl, format, layout, audience and productId. The description reinforces the 'brief' semantics ('using only the facts in it') and the draft nature, but adds nothing for the nested audience object or the enum options, so baseline 3 is correct.

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 ('has the workspace's AI model write a newsletter from a brief') plus the persistence outcome ('saves it as a draft'). It is clearly distinguishable from siblings like create_newsletter or update_newsletter by being AI-generated rather than manually authored, though it never names those siblings explicitly.

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 a prerequisite ('Needs the AI set up in Settings → AI') and an outcome constraint ('Nothing is sent'), which frames the intended workflow: produce a draft for human review. However, it never says when to choose this over create_newsletter / write manually, nor what to do next (preview, schedule, update) once the draft exists.

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