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

Annotations already declare non-readonly, idempotent, non-destructive. The description adds real value beyond that: it clarifies the output is an ordinary draft that is not auto-sent, and that incomplete variable substitution fails rather than producing a partial letter. It omits that a content_policy_acknowledged=true and an idempotency_key are mandatory, which are the two most likely failure modes.

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 with the 'Mail merge' category marker first and the key precondition second; nothing is redundant or 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?

For a 9-parameter nested-schema mutation tool with no output schema, the description leaves gaps: it does not say what comes back (a draft id/object) nor flag the two required fields that are not obvious payload data. Adequate to understand the concept, thin for calling it correctly on the first attempt.

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 only 22%, so the description needs to compensate. It covers the template source, variable substitution and the recipient/sender inputs, but says nothing about template_id, sender_id/recipient_id precedence over inline addresses, the idempotency_key, or the content policy flag.

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 ('create one ordinary draft from a template') and front-loads the 'Mail merge' framing, which distinguishes it from create_draft. It does not explicitly name create_draft as the non-template alternative, so sibling differentiation is implied rather than stated.

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?

The phrase 'Missing variables are rejected' plus 'Proof, quote and send it like any draft' implies the downstream workflow and a precondition, but there is no explicit guidance on when to pick this over create_draft vs create_batch, and no stated prerequisites such as the content policy acknowledgment.

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