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

Annotations already declare readOnly/idempotent/non-destructive, and the description adds substantial context beyond them: server-side masking of emails and phones, free-of-charge sampling, deterministic repeats, the empty-result threshold at 25, and the PECR/corporate-subscriber default that silently narrows the email audience. That last point is a hidden scoping behaviour an agent could not infer from annotations or schema.

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 most decision-relevant facts (what comes back, the 25/10 limits, the empty-result threshold) are front-loaded, and the workflow pointers follow. The trailing legal-basis sentence is long and partly boilerplate for an invocation decision, but it does carry the PECR exclusion, so it is not wasted.

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, yet the description still summarises the row shape and, more importantly, flags the surprising empty-result condition and the deterministic repeat behaviour. Combined with the workflow ordering and the corporate-subscriber default, an agent has everything needed to call this correctly and to interpret a zero-row response.

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 75%, so the schema already carries most parameter meaning; the description restates the audience-or-count-id choice and the 25-row ceiling without adding format or edge-case detail. It says nothing about the shuffle parameter, which is the one parameter whose behaviour (new seed, counts against rate budget, logged) would have benefited most from narrative explanation—though the schema does cover it.

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 (returns masked sample rows for an audience or count id) plus the shape of each row: verification badges, match reason and provenance card. It also implicitly separates itself from the paid retrieval path by pointing at quote_order, so an agent can distinguish it from fetch_order_records.

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 explicit sequencing: 'Use it after count_audience' to show the user what records look like before quoting, and 'Call quote_order next to price the records.' It also states the condition under which it yields nothing useful (count under 25), which is a de facto when-not-to-use rule.

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