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

Annotations already cover read-only, idempotent, non-destructive, and open-world, so the safety profile is done. The description adds real behavior: scoring is weighted across only the checks that apply, the count of evaluated checks is reported, and page-speed consent triggers an external Chrome UX Report lookup on the busiest landing pages. It doesn't disclose cost, latency, or failure modes for that external call.

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 'One number:' and builds outward to what is returned, then the consent caveat, then read-only. Sentences are dense but each carries content. The trailing 'Tools revision: d80f65c0.' is build metadata that adds no agent-relevant information and slightly dilutes the ending.

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?

Return values are safely covered by the output schema, so the description needn't describe the shape. However, with four parameters at 0% schema description coverage, an agent has no guidance on the lookback window or what summary versus full detail yields, which is a material gap for a scoring tool. The behavioral and purpose framing is otherwise complete.

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 description coverage is 0% across four parameters, so the schema carries no meaning and the description must compensate. It never mentions 'days', 'detail' (summary vs full), or 'render' (json/markdown/both), leaving the two enums and the lookback window wholly unexplained. Only an indirect hint exists in that a per-category breakdown may relate to detail level.

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-free but precise resource and output: a weighted 0-100 health score over the applicable subset of 75 audit checks, plus letter grade, per-category breakdown, and evaluated-check count. It explicitly differentiates itself from fuller audit reporting with 'No findings: the cheapest check of how the account is doing', which separates it from full_audit.

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?

The phrase 'the cheapest check of how the account is doing' clearly signals when to reach for this over heavier audit tools, and the page-speed consent caveat gives a usage precondition. It stops short of naming full_audit or quality_score_breakdown as the alternative to use when findings are wanted, so the routing is implied rather than explicit.

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