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

Beyond the readOnlyHint annotation, the description discloses the core behavioral contract: `text` 'selects an archetype from a closed catalogue' and is 'never used to construct anything,' the tool deliberately answers 'with no answers at all' using documented defaults, and a location constraint 'is reported as unmet, because this tool prices ONE region.' It even reveals the failure mode — the reply says so and lists what the catalogue covers instead of inventing a result — which is exactly the kind of surprise-avoidance transparency that matters.

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 description is long, but every paragraph contributes operational value: usage trigger, matching semantics, reply design rationale, parameter-by-parameter behavior, and display guidance ('Show it'). The core purpose is front-loaded in the first sentence, and even the rhetorical closing line ('a number with no visible shape behind it is not something a person can check') earns its place by explaining why the agent must surface the architecture.

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?

Exceptional for a 6-param tool with zero schema parameter descriptions and no output schema: it covers the reply structure (`open_questions` ranking, `how_we_got_the_numbers`, `architecture`), the single-region limitation, the closed-catalogue matching, and the no-match failure mode. The only real gaps are the never-defined `term` parameter and the only-implied `region` parameter — modest omissions against an otherwise complete picture.

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?

With 0% schema coverage, the description carries the full burden and does so deeply: `text` is defined as keyword-based catalogue matching, `answers` takes 'human units... not cloud units,' `constraints` comes with an explicit JSON shape, and `archetype` is sourced from a sibling tool. However, `term` (default 'OnDemand') is never mentioned at all, and `region` is only implicitly defined via the one-region limitation, which keeps this from a 5.

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 opens with a precise action — 'Price a workload described the way a person would describe it' — and grounds it with concrete examples ('a website with a database for my shop'). It clearly distinguishes this tool from siblings by positioning it as the natural-language entry point and naming `list_cost_building_blocks` as the structured alternative, so an agent can tell it apart from `estimate_cost` and `price_lookup` without opening their schemas.

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 an explicit trigger ('Use this when the user does not know cloud'), states the exclusion ('you should not name any on their behalf before calling this'), and names alternatives directly: pass `archetype` from `list_cost_building_blocks` and use `where_can_this_run` for location-placement questions. It even provides stopping guidance — 'ask the top one first and stop when the range is tight enough' and 'Do not ask all of them' — leaving nothing to inference.

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