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

Annotations already declare read-only, idempotent, open-world behavior, and the description still adds meaningful context beyond them: runtime (5 to 30 seconds), result caching including how to bypass it (fresh: true with a 5-minute threshold), the structure of the return payload, and the explicit limits of an automated scan. That is exactly the extra context the lower bar with annotations asks for.

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 usage trigger, then behavior, then a single-constraint close. It is fairly dense at roughly 140 words, but every sentence carries information an agent needs; the only mild redundancy is restating the return payload when an output schema already exists.

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?

For a 6-parameter scan tool with annotations and an output schema, the description covers trigger, duration, caching, scope limits and per-call cardinality. Nothing an agent needs in order to call it correctly 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?

Schema coverage is 100%, so the baseline would be 3, but the description adds real semantics by tying `fresh` to the caching rule ('A recent result of the same page may be reused (reused_until); fresh: true scans it again') and by stating the one-page-per-call constraint. The remaining parameters (viewport, include_rgaa, max_examples, include_best_practices) are left entirely to the schema.

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 ('Loads one public web page in a real browser and runs axe-core against WCAG 2.2 level AA'), which distinguishes it cleanly from sibling check_eaa_scope and from 'reading the page source'. An agent knows exactly what it produces 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?

Opens with an explicit trigger ('when a user asks to audit, check or test the accessibility of a website or web page'), names the standards and page types, gives a prefer-it-over alternative ('Prefer it to reading the page source: it tests the rendered page'), and closes with a cardinality rule ('One page per call: call it again for each other page').

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