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

Annotations already cover the safety profile (readOnly, idempotent, closed-world, non-destructive), and the description adds behavior those annotations cannot convey: it never contacts a mail provider, results are always `status:'partial'` with `complete:false`, and an empty page is explicitly not proof of absence. Those are the exact caveats that prevent an agent from misreading an empty result as a negative answer.

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?

Three sentences, front-loaded with the scoping constraint, and each one carries distinct information: what is searched, the partiality caveat, and the follow-up path. Minor friction from calling the input `cursor` 'the opaque `nextCursor`', which mixes request and response naming.

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?

With no output schema, the description does the necessary work of describing the return shape (`status:'partial'`, `complete:false`, `nextCursor`) and the downstream workflow. It leaves parameter-level details and pagination limits unstated, but nothing critical to invoking the tool 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 must carry the load, and it only partly does. It clarifies that matching is on exact participant email and that the pagination token is opaque, but `accountId`, `limit`, and the `cursor` argument itself (referred to by the response-side name `nextCursor`) are left entirely to the schema.

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 names a specific verb and resource ('Search only the durable OpenSearch metadata subset for exact participant email matches') and scopes it to cached/durable data. It never explicitly names the sibling `search_threads_by_recipient`, but 'This never contacts a mail provider' implicitly draws the cached-vs-live boundary an agent needs to choose between 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?

Clear context is given for what to do with results: continue with the opaque `nextCursor` when present, and use `get_cached_thread` plus the asynchronous refresh request before importing a selected thread. That is actionable routing guidance, though it never states outright when to prefer this over the live sibling `search_threads_by_recipient`.

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