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

Annotations already declare readOnly/idempotent/non-destructive, but the description adds context annotations cannot carry: this system has no send capability (CRM-14), so nothing in the list was sent by the tool, and status="sent" records what a person sent and reported back. That prevents an agent from mistaking 'sent' for an automated delivery and is exactly the kind of domain caveat that matters here.

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?

The purpose and the actionable warning are front-loaded in the first two sentences, and every sentence afterward contributes a distinct semantic point. The CRM-14 parenthetical and the trailing clause about discarded reasons are slightly rambling, but no sentence is 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 a read-only filtered list with full schema coverage and no output schema, this covers purpose, call timing, and the meaning of the non-obvious enum values. Minor gaps remain — result ordering and what a returned record contains beyond contact/account are not described — but nothing needed to invoke it correctly is missing.

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 mechanics of limit/status/account_id/contact_id are already documented, but the description adds real semantic depth to the status values beyond the schema's one-line 'Default draft' note — explaining what sent and discarded actually represent in this system. That meaning is not recoverable from the enum alone.

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?

First sentence names the resource ('messages written but not sent') and its scope ('with the contact and account behind each'), so the agent immediately knows this is the read/list counterpart to crm_draft_message and crm_update_draft rather than a write tool. The follow-up sentence ties it to the drafting workflow, which further separates it from the create/update siblings.

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?

'Check before writing a second one to somebody who already has one waiting' gives an explicit trigger for calling it ahead of crm_draft_message, and the note that discarded reasons are 'the most useful thing ... when you write the next one' reinforces the pre-drafting use case. It doesn't state when not to call it or name alternatives by tool name, keeping it just short of a 5.

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