apps_interface_input

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

Annotations already declare destructiveHint=true, non-idempotent, and open-world, so the safety profile is covered. The description adds non-obvious behavior: delivery only confirms sent/dispatched events rather than a verified outcome, and verification requires screenshots or App state. That is meaningful context beyond the structured hints.

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?

It is front-loaded with the action verbs and then layers constraints, with each sentence carrying distinct information (prerequisite, exclusion, unit semantics, verification caveat). It is dense and slightly long, but no sentence is 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?

For a high-complexity, nested-event, no-output-schema tool, the description covers modality, prerequisites, coordinate/unit semantics, exclusions, and the limits of delivery confirmation. An agent has what it needs to invoke and verify correctly.

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 50% and instanceId/generation are undocumented, but the description compensates for the input-critical parameters by specifying that pointer coordinates are normalized App coordinates and scroll deltas are in CSS pixels, which the schema does not state. The generic instanceId/generation params remain unclarified.

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 specific verbs (click, drag, scroll, type, send key/chord) and the resource (the active device App interface), so the core action is unmistakable. It also carves out the guide_start event subtype. However, it never distinguishes this tool from the sibling apps_preview_input, leaving overlap around preview inputs unresolved.

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?

It states when to use guide_start, what guide_start does not do (does not approve access or I/O), that other host menus and permission dialogs remain human-operated, and that apps_input_targets must be read first. Prerequisites, exclusions, and the routing to the targets tool are all explicit.

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