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

Annotations cover the read-only/idempotent profile, but the description adds substantial context beyond them: index fallback order (district → local authority →national), the `index_unavailable` carry receipt and no-adjustment outcome, UK-only scope, and a concrete cost of 10 credits (10p). It also flags that the -10% relisting penalty is an assumption, not a measured discount — exactly the kind of caveat an agent cannot get from structured fields.

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?

It is front-loaded with the payload and every sentence earns its place, but the central sentence is a ~60-word run-on with stacked parentheticals (index fallback, `index_unavailable`, relisting caveat) that take effort to parse. Splitting that block would cost nothing in content and improve scanability.

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 shape needs no explanation, yet the description still summarises what each comp carries. Combined with cost, jurisdiction, keying, and index-fallback behaviour, an agent has everything needed to call this correctly and interpret the result.

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 0%, so the description carries the full burden and does: `relisted_uprns` is reframed as an explicit sensitivity switch with a stated -10% magnitude, `lookback_months` gets an enforced 12–120 range, and `uprn` is qualified as a UPRN rather than an address with the resolution path spelled out.

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?

It names a specific resource (same-building comparable sales with registered price, date, time adjustment, relisting penalty, IQR fence) and explicitly contrasts with `valuation_full`, which 'gives the value itself'. The distinction from `property_comps`/`valuation_comparables` is not drawn by name, but the 'evidence behind a valuation' framing is precise enough to separate it from every other comps sibling.

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?

It tells the agent when to reach for it ('Use it to show WHY a value is what it is'), names the alternative for the value itself, warns to pass `relisted_uprns` only for a deliberate sensitivity rather than a default, and spells out the prerequisite chain (`address_autocomplete` free → `address_resolve` → UPRN).

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