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

Annotations only say readOnlyHint=false, destructiveHint=false and idempotentHint=false; the description goes well beyond that by disclosing the side effect (a SavedSearch row is created on the user's account), the required OAuth scope `listings:read`, the delivery mechanism (email on new matching listing or price drop, instant or daily digest per user preferences), and the per-user 50-active-alert cap. These are exactly the operational facts an agent needs before mutating account state.

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 sentence is front-loaded and the parenthetical about read-only data vs. an actual write is a high-value clarification, not filler. The example sentence is the longest element and is only partially earned since the filter fields are already in the schema, but nothing is redundant with the annotations.

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?

An output schema exists, so return values need no explanation, and the description covers the write semantics, auth scope, delivery behavior, and quota for a 9-parameter mutation tool with no required fields. The one gap is the non-idempotency consequence (whether saving the same filter twice creates duplicate alerts), which matters given idempotentHint=false.

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% and six parameters (brand, model, minPrice, maxPrice, condition, category) have no schema descriptions, so the description must compensate. The examples implicitly map brand/model/city/price/category onto the filter, and "instantly or in a daily digest" clarifies the alertFrequency intent (which does have a schema description anyway). It still adds no meaning for `name` or for minPrice vs. maxPrice interaction, leaving some of the gap unfilled.

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?

"Subscribe the user to email alerts for Cyclesite listings matching a filter" states a specific verb (subscribe) and resource (saved search alert), and the noun "subscribe" cleanly separates it from the read-only search/search_bikes siblings. An agent can tell this creates a persistent alert rather than running a query.

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 two quoted examples ("let me know when a Trek Domane SL 6 in Manchester under £2,000 appears", "alert me to any e-MTB drops below £2,500 in Yorkshire") make the intended trigger situation concrete, and the 50-alert cap gives a precondition. It stops short of explicitly naming when to prefer this over search_bikes (one-off lookup vs. ongoing monitoring), so the exclusion is inferable rather than stated.

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