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

Annotations only declare the non-read-only, non-destructive, non-idempotent profile; the description goes well beyond that by specifying merge/partial-update semantics (unnamed settings keep their stored value), ownership enforcement with the 404 failure mode, the permission scope needed, the sign-in side effect for unconnected members, and where the change is recorded. The prompt-injection caution about member-written text is unusual but genuinely useful behavioral context for an agent handling site data.

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 core mutation semantics and the ownership/permission constraints are front-loaded in the first sentences, and each sentence carries a distinct fact. The closing sentence about member-written text is generic boilerplate that is only loosely tied to this specific tool, costing a little tightness.

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 two-parameter mutation with a nested object, no output schema, and partial annotations, the description covers auth, ownership, partial-update behavior, and side effects well. It never indicates what the call returns or how to confirm the new alert state, which is 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 description coverage is 100% and the nested alerts object documents each toggle and the enum values, so the schema already carries parameter meaning. The description adds only the emphasis that unspecified keys are preserved, which largely restates the schema's own "Leave a key out and it keeps the button's default" note, so baseline 3 is right.

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?

"Change a watch's alerts" states a specific verb (change) and resource (a watch's alert settings), and the follow-on sentence clarifies it mutates the Watch button's own toggles rather than creating or removing the watch. That distinguishes it from watch_item/watch_person/unwatch, though those siblings are never named explicitly.

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 clear preconditions and context: only the member's own watches (404 otherwise), requires the watch:write permission, and triggers a sign-in prompt if the member has not connected the integration. It does not explicitly compare against alternatives such as unwatch or set_notification_schedule, so the agent must infer which sibling to pick.

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