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

With no annotations provided, the description carries the full burden of behavioral disclosure. It does disclose key behaviors: both fingers ride a vertical axis through the centre, fingers are auto-clamped on-screen, coordinates are screen points, and it requires an active session. It additionally explains the scale ratio semantics (2 doubles, 0.5 halves) which is genuinely useful behavioral context. However, it does not disclose whether the gesture is a discrete event or continuous, or behavior on failure (e.g., no session), which a mutation-like tool with zero annotations could reasonably disclose.

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 description is tight, roughly 4 sentences, dense with useful information and zero filler. Front-loaded with the core action ('Two-finger pinch-to-zoom centred at (x,y) on an iOS device'), then explains scale semantics, then finger behaviour, then use cases, then coordinate system, then a requirement. Every sentence earns its place.

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 gesture tool with complete schema coverage, no output schema (thus no return-value expectations to document), and no annotations, the description is well-rounded. It covers purpose, scale semantics, finger positioning behaviour, use-case applicability, coordinate system, and session requirement. The main gap is that with no annotations and no output schema, it could mention return value or feedback behaviour more explicitly, but this is a minor omission for a gesture action.

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% for all 5 parameters, so the baseline is 3. The description adds genuine value beyond the schema: it explains scale semantics ('~2 doubles, ~0.5 halves'), the vertical axis behaviour of the fingers, and the auto-clamp behaviour. It also clarifies that coordinates are screen points. This exceeds the baseline by adding meaning beyond what the schema's field descriptions provide, though it doesn't document durationMs explicitly beyond the default in 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?

The description provides a specific verb+resource: 'Two-finger pinch-to-zoom centred at (x,y) on an iOS device'. It clearly distinguishes from siblings (device_swipe, ios_swipe, ios_drag_drop, device_scroll etc.) by being the only gesture that describes multi-finger pinch semantics. The precise scale semantics (scale > 1 in, < 1 out) add specificity beyond a generic purpose statement.

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?

The description names explicit use cases ('Maps, Photos, Safari, camera') which gives clear context on when to use it. It also states a requirement ('Requires an active iOS automation session'). It does not explicitly name alternative sibling tools or state exclusions, though the iOS-native nature is clear from the name and it points users to the ios_* family implicitly. Lacking explicit 'use X instead' guidance prevents a 5.

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