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

Annotations already declare the mutation, destructive, and idempotent profile, but the description goes well beyond them: it names three specific 409 failure modes with their conditions, explains that changes propagate to the AI only from its next reply, and describes what the response's `message_weight` and optional `notice` mean and that the notice should be relayed to the user. This is exactly the extra operational context annotations cannot carry.

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 line is front-loaded, the conditional rules for `active` and `type` follow in a logical order, and the response-handling note closes it out. Sentences are dense with facts (error codes, caps, propagation timing) rather than filler; the only mild redundancy is that the `type` media precondition also appears in the schema.

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 an 8-parameter mutation tool with no output schema, the description does the important work: it explains failure conditions the agent may need to handle and describes the return payload's `message_weight` and `notice`. What is missing is a brief note on which fields are updatable as a set or what happens when no fields are supplied, which would close the remaining gap.

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 75%, so a baseline of 3 applies, and the description earns the bump by tying `active` and `type` to concrete preconditions and error codes rather than restating the schema. It leaves `priority`, `when_to_send`, `name`, and `description` entirely to the schema, which is acceptable given the high coverage but keeps it short of a 5.

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 opens with a specific verb+resource ('Change an AI attachment'), and 'Changes the fields you send' clarifies that this is a partial update rather than a replace. It doesn't explicitly distinguish itself from siblings like create_attachment, delete_attachment, or get_attachment, but the title and first line make the intent unambiguous.

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?

Usage conditions are given, but only for specific fields and error paths: `active: true` requires stored media and room under the 100-active cap, and `type` is editable only while no media exists. There is no statement of when to reach for update_attachment instead of delete_attachment plus create_attachment, or when an update is preferable to leaving the attachment untouched, so the when-to-use-vs-alternatives guidance is implied rather than explicit.

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