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

Annotations (readOnly, idempotent, non-destructive) cover the safety profile, and the description adds substantial non-obvious behavior: the cursor is an event id not a timestamp, late rows take higher ids, empty pages must be checked against coverage.event_types, delisted effectively has no coverage, price change source semantics, and that null means unknown rather than zero. It also gives cost (5 credits) and UK-only scope. The only reason it isn't a 5 is that a couple of behavioral edges (pagination/limit behavior) are left implicit.

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?

Front-loaded with the core mechanism, then layers caveats in a logical order. It is on the long side and uses emphasis/bold for the delisted warning, but nearly every sentence carries decision-relevant information rather than filler.

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?

Given a read-only feed with 7 params, zero schema coverage, and an output schema already present, the description supplies the missing piece: coverage limitations an agent must respect before trusting results, plus the scoping precondition and sibling routing. Nothing essential for correct invocation 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?

With 0% schema description coverage, the description carries the load and largely succeeds: it defines the meaning of cursor (event id, from previous call), since, uprns, and postcode_districts. However it references the cursor as 'next_cursor' while the schema field is 'cursor', and it never explains the limit or channel parameters, so compensation is thorough but not complete for all 7 params.

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?

Opens with a specific verb+resource: 'What changed since you last looked — the portal change feed.' It names the cursor-driven feed precisely, and the surrounding text makes clear this is a change/event feed rather than a static history tool (property_history, listing_history). An agent can place it against siblings without opening the schema.

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?

Explicit when-to-use mechanics: pass next_cursor for incremental reads, or supply a scope (postcode_districts/uprns) or since date on a first call because unbounded reads are refused. It also routes to siblings ('call address_autocomplete (free) to narrow, then address_resolve to get one') and states the cannot-answer cases, which is genuine when-not guidance.

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