send_message

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

Annotations already declare destructive/openWorld behavior, but the description adds important operational context: host approval is required, confirm is never set by the client, calls are audited, and an idempotency key controls retry behavior. These are valuable behavioral details beyond the annotations, and they do not contradict them.

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 description is short and front-loaded with the action, then the destructive approval requirement, then retry semantics. Every sentence carries useful information.

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 send operation with no output schema, the description covers the critical safety and retry behavior well. It is still missing return-value expectations and specifics for content/conversationId, which keeps it from being fully complete.

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 50%, and the description reinforces confirm while clarifying that same idempotencyKey retries the same send. It does not explain content or conversationId semantics, leaving those to the agent's inference from the tool name and schema.

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 gives a specific verb and resource: "Send a reply." That clearly distinguishes it from sibling list/get tools, but it does not explicitly differentiate it from nearby draft/reply tools such as save_draft or approve_ai_reply.

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 states the key prerequisite that the host must approve the call and that confirm must already be true. However, it does not say when to use this instead of save_draft or approve_ai_reply, so usage selection is only implied.

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