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

Goes well beyond annotations: discloses asynchronous execution, expected duration ('a few minutes'), and deduplication behavior (returns the in-progress scan id instead of starting a duplicate), which concretely explains the idempotentHint. Also clarifies the mutation is non-destructive.

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?

Three tightly written sentences, front-loaded with the action, then async behavior, then the dedupe caveat. Every sentence carries actionable information with no 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?

With no output schema, the description does well by directing the agent to get_scan_status for results and describing the duplicate-id return case. It stops just short of explicitly stating that a successful new scan returns an id, which is a minor omission for an async starter tool.

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?

Only one parameter and the schema already fully documents it, including how to obtain it via list_domains. The description adds only 'one of the user's domains', which is consistent with but not substantially beyond the schema; baseline 3 applies at 100% coverage.

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 and resource ('Start a new SEO/UX scan (crawl)') plus scope ('for one of the user's domains'). It clearly distinguishes itself from the sibling get_scan_status, which is named as the follow-up tool.

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?

Explains the async nature and explicitly routes progress-checking to get_scan_status, and covers the duplicate-scan case. It lacks explicit 'when not to use this' guidance (e.g., domain must already exist via list_domains), though that is partly implied by the schema.

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