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

With no annotations provided, the description carries the full behavioral burden and delivers: it discloses the snapshot-at-instant timing behavior with measured evidence (4 issues mid-animation vs 5 once settled), the platform engine limitation (can only audit a whole app, so element filtering cannot surface new findings), the version failure mode, and session auto-start. Every significant behavioral trait an agent needs to interpret results is disclosed.

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?

Long but dense — every sentence carries non-redundant information, organized with clear signposts (TIMING:, Requires) and front-loaded with purpose and prerequisites before behavioral caveats. No filler, no restatement of the schema's parameter descriptions.

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?

Complete for its complexity: no output schema exists, but the description specifies what each issue carries (label, type, screen-point rect) and how to act on it; no annotations exist, but behavior, preconditions, failure modes, and version requirements are all covered. An agent can decide when to call it, what to pass, and what to expect in response.

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% and the schema itself is unusually detailed (rect precedence and centre-rule, element resolution and failure mode, per-enum expansion of audit types). The description still adds value beyond the schema: the performance hint that auditTypes narrows the run and is 'much faster on dense screens', plus the actionable link that a finding's rect centre is directly tappable with ios_tap.

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 names a specific verb and resource — 'Run an accessibility audit on the CURRENT screen of an iOS device' — and identifies the exact engine (Apple's XCTest audit engine, same as Xcode's 'Audit for Accessibility'). The platform scoping ('iOS device') and the enumerated finding types (contrast, tap targets under 44×44pt, missing labels, traits) clearly differentiate it from the android_accessibility_audit sibling.

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?

Explicitly sequences the workflow: 'navigate to the screen you care about FIRST (ios_tap_by_label / ios_navigate_url), then call this', and gives an explicit when-not condition — running right after launch or navigation under-reports while the UI animates ('Let the screen settle first'). It also states preconditions and failure modes: iOS 17+ (errors on older) and an active iOS automation session (auto-starts if needed).

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