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

Annotations already declare readOnlyHint/idempotentHint=true and destructiveHint=false, so the safety profile is covered. The description layers substantial disclosure beyond that: headless-chromium requirement, disabled-on-hosted behavior with concrete measurements, scroll-settle/video-play semantics (fires whileInView, plays preload=none videos), and the three-tier evidence tagging. No contradiction with annotations — an audit tool reads and reports, consistent with readOnly=true.

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?

The best information is front-loaded into the first sentence (render-and-capture, live URL, viewports×themes), with operational details and limitations following in a logical cascade. It is dense, but every sentence carries unique value — cost, environment, alternatives, output contract — making the length justifiable for a tool this complex.

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 owns the return contract and fulfills it: 'Every finding is tagged confirmed | likely-artifact | inconclusive with its evidence, ranked by severity'. It covers prerequisites (headless chromium), environment constraints (disabled on hosted with timing evidence), resource cost (~95–120s), and a fallback path (audit_page for static HTML), leaving no material question answered for the caller.

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 100%, so baseline is 3; the description meaningfully adds by explaining why behaviors matter (interactions are 'diffed against baseline to catch hover/click white-wash', scroll_settle 'fires reveal-on-scroll/IntersectionObserver content'). Only minor gap: the compact vs includeScreenshots false-default tension and their combined use is implicit rather than explicitly called out.

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 concrete verb+resource ('Layer 0 render-and-capture audit') and specifies exact scope: LIVE URL, viewport×theme matrix, scroll-settling, optional interactions, real pixels + rendered DOM. Actively differentiates from siblings by naming the exact class of bugs it catches ('cropped images, blank videos, hover white-wash, sliced exports, hidden-on-mobile') that are 'invisible to HTML-string/geometry audits', cleanly separating it from audit_page, audit_contrast, and audit_layout.

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?

Provides explicit when-to-use ('This is the tool that catches real-world visual nits...'), an explicit alternative for static input ('pass the page HTML to audit_page here for a static grade'), and deployment-specific instruction ('Run it locally instead (npx raven-mcp) where there is no request deadline'). Measured cost data (95s cheapest, >120s defaults) lets an agent reason about whether the call is worth the budget.

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