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

With annotations already declaring openWorldHint=true, idempotentHint=true, and destructiveHint=false, the description still adds real behavior: the fetch-then-LLM pipeline and a 24h cache. What it does not explain is why readOnlyHint is false – whether calling creates a persisted audit record, consumes LLM quota, or has cost/latency implications. For a tool flagged as non-read-only, that is a meaningful unexplained trait.

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?

Two tightly packed sentences: the capability and its output come first, then the scope exclusion, then the cache attribute. No sentence is filler, and the most decision-relevant constraint (not a technical audit) is placed early.

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?

The tool is simple (one required parameter) and has an output schema, so return-value detail is unnecessary. The description covers the analysis dimensions, the scoring range, the exclusion set, and caching – everything an agent needs to call it 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?

The single url parameter has 100% schema coverage and already states it must be a public page, so the schema carries the burden. The description adds only indirect meaning by saying it fetches "the page's title and main content," which hints that non-rendered or auth-gated pages may yield poor results, but does not state that explicitly.

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 gives a specific verb and resource ("Run an AI-generated SEO audit on a URL") and spells out the analysis surface: title, meta description, headings, content quality, internal linking. It also draws a clear boundary against technical/performance auditing, so an agent knows what it is and is not. It never names a sibling such as list_page_audits, so the distinction between generating an audit and retrieving stored audits is left implicit.

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?

There is explicit when-not guidance: it is a qualitative content assessment and does not return Core Web Vitals, load time, or Lighthouse metrics. The 24h cache note also tells an agent that repeat calls are cheap, implicitly discouraging redundant re-runs. It stops short of routing to alternatives (e.g. get_metrics for technical data) or stating prerequisites.

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