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

Annotations declare destructiveHint=true, idempotentHint=true and openWorldHint=true, but the description goes well beyond them: it says a watch is created, a first check runs immediately, recurring checks are scheduled, changes surface as alerts and may be emailed, identical settings reuse a paused watch without resuming it, and request_id deduplicates. That is exactly the operational context an agent needs before invoking a stateful 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?

The purpose is front-loaded in the first sentence, and every subsequent sentence adds operational detail rather than filler. The inline `what:`/`how_often:` enumeration makes the middle dense and slightly run-on, but no sentence is dispensable.

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 five-parameter, stateful tool with no output schema, the description covers creation, immediate first run, scheduling, notification channels, reuse semantics, idempotency, and the returned watch id used for later management. Nothing material an agent needs to call it correctly is missing.

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 0%, so the description carries the full burden, and it does: it enumerates the accepted `what` values, the accepted `how_often` intervals with the plan-based floor, the role of `under_price` as a target for price watches, and the deduplication meaning of `request_id`. Only the obvious `link` parameter is left unstated, which is acceptable.

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?

'Keep checking a page and tell the user when it changes' gives a specific verb and resource, and the follow-up sentence enumerates the five watch modes (price, stock, change, jobs, account). The agent can distinguish this from one-off siblings like see_page, read_site, or check_watch_now because the description stresses a persistent, recurring watch.

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?

The `what` enumeration effectively tells the agent which situation maps to which mode, and the reuse/idempotency sentences clarify when a new watch is or isn't created. It stops short of explicitly naming alternatives such as using read or see_page for a single check, leaving the recurring-vs-one-shot distinction implicit.

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