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

Beyond the annotations, the description discloses the critical safety fact that nothing is ever sent and that organization sending policy is applied by a human reviewer, plus concrete idempotency behavior ("Pass the same idempotency_key when retrying; a repeat returns the original draft"). It leaves the mutual-exclusivity of conversationId/clientId and the channel default unspecified, which is a real behavioral gap for a write tool.

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, each earning its place: the core action, the no-send guarantee, then the identifier and idempotency mechanics. The most decision-relevant fact ("never sends") is front-loaded right after the action.

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 draft-creation tool with no output schema, the description covers the essential risks and retry contract, and "a repeat returns the original draft" hints at the return value. It could go further by stating the draft identifier returned and resolving the conversationId/clientId either-or rule.

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 20%, so the description must carry the load; it clarifies the relationship between conversationId and clientId and explains retry semantics for idempotency_key. It says nothing about body (required, 1-4000 chars) or channel (sms/email enum, defaulting behavior), so roughly half the parameters remain undocumented anywhere.

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 description uses a specific verb and resource ("Writes a message into a client conversation's draft box") and immediately distinguishes itself from the sending siblings (send_text, send_email) by stating "It never sends." An agent can tell exactly what this creates and what it does not.

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?

It gives clear usage context: draft-only, a human presses Send, and which identifier to supply (conversationId, or clientId to use that client's conversation). It does not name edit_draft or redraft_lead_step as the tools for modifying an existing draft, nor state what happens if neither id is supplied, so routing guidance is strong but not exhaustive.

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