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

Annotations only say readOnlyHint=false, destructiveHint=false, openWorldHint=true — which could lead an agent to underestimate the impact. The description adds the crucial disclosure that this 'really delivers to the other person', i.e. an irreversible external side effect, plus the sending identity and account routing. That is exactly the value-add the annotations cannot provide.

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?

Two short sentences with zero filler; the delivery/impact warning is placed where it will be read before the confirmation requirement, which is the right ordering for a consequential 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 3-param mutation tool with no output schema and open-world reach, the description covers identity, target scope, real-world effect, and a safety precondition. What's missing is minor: return value (message id) and the source of conversation_id, both of which an agent could infer from siblings.

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 33% — only thread_id is documented, and the schema itself explains it. The description adds that conversation_id must reference an existing conversation and that text should be user-confirmed, but it never says where conversation_id comes from (e.g. list_conversations) or how thread replies behave. Partial compensation for a real gap.

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 (send), resource (a message), identity (as the user), and target (an existing conversation via its owning account). This clearly separates it from read-only siblings like read_conversation, search_messages, and list_conversations without needing to name them.

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 concrete precondition ('confirm the text with the user first') and a scope constraint ('into an existing conversation'), which tells the agent this is not for starting new threads. It stops short of naming an alternative tool or explicitly saying when not to use it, so it falls just below the top band.

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