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

Annotations already declare readOnly, idempotent, closed-world and non-destructive, so the safety profile is covered. The description adds real context beyond that: newest-first ordering, that results come from one of the panel's tabs, the pagination contract ("Pass next_cursor back as cursor"), and a definition of the status/email_status/signed vocabularies. It stops short of stating the page size or what next_cursor looks like when exhausted.

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 opening sentence leads with the resource and the returned fields, then moves to ordering, filtering and pagination — a sensible front-loaded order. It is dense but each sentence carries information; the only soft spot is that the status/email_status/signed explanation reads like output documentation without ever flagging that these are returned values, which takes a second read to disambiguate from input parameters.

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?

There is no output schema, so describing the returned fields and the status vocabulary is genuinely load-bearing here and the description does it. Pagination, tab filtering, search and client scoping are all covered. The gap is `limit` (default and ceiling unstated) and no note on what happens when the result set is empty or how many records a page holds.

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 80% and the existing schema descriptions for cursor, search, approval and client_id are essentially restated rather than extended. Most of the descriptive text explains returned fields (status, email_status, signed) rather than the inputs, and `limit` is never mentioned in either place despite having a default of 40 and a max of 100. Baseline 3 is the fair score when the schema already carries the parameter semantics.

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?

States a specific verb and resource ("Lists estimates") plus the exact shape of what comes back: number, client, total, expiry, status, approval, emailed/signed. Ordering ("newest first") and the tab scope are given. It does not explicitly contrast itself with get_estimate or list_invoices, so an agent still has to infer that this is the bulk/panel view rather than a single-record or invoice fetch.

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 is implied through the filter narrative — pick a tab, optionally add a search term or restrict to one client — which is decent guidance for choosing parameters. However, there is no explicit when-to-use-this-instead-of statement, e.g. that get_estimate should be used to fetch a single estimate's full detail or delete/update_estimate for mutation. The agent must derive the boundary with siblings itself.

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