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

Annotations already declare destructiveHint=true, idempotentHint=false, and openWorldHint=true, so safety is covered. The description adds real value beyond that by disclosing that the batch halts at the first failure and what the result contains (the screen where it stopped, unless the screen couldn't be read or not in time) plus the anti-retry rule, though it says nothing about the destructive scope of individual actions or rate/time limits beyond the batch cap.

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?

Three sentences, front-loaded with the core capability and the action list, then failure behavior, then the operating rule, with no wasted filler. It loses a point because the 'screen where it stopped, unless the screen couldn't be read, or not in time' clause is restated almost verbatim in the observe parameter description, and the trailing ellipsis is slightly informal.

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 complex multi-variant action tool with no output schema, the description adequately covers what the tool does, the batch size, stop-on-first-failure behavior, and what the result reveals of the screen. It leaves unstated the relationship to observe as the recommended preview step before batching and the fact that actions carry full destructive/open-world scope, but nothing essential for calling 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 description coverage is 100%, with maxItems, action variants, enum values, and the observe enum all documented in the schema. The description's 'up to 20 actions' merely restates maxItems and adds no targeting syntax, coordinate units, or targeting-mode semantics beyond what the schema already provides, so the baseline 3 applies.

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 gives a specific verb and resource ('Run up to 20 actions in order') and enumerates the concrete action set (tap, type, key, swipe, scroll, back, home, open_app, open_url, wait_for), so an agent immediately knows what the tool does. It only gestures at sibling differentiation via 'observe first' rather than naming the observe tool or drawing a clean line between acting and observing.

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 guidance 'Never repeat an action whose outcome is unknown; observe first' tells the agent how to sequence this tool with verification and implicitly routes to the observe sibling. There is still no explicit statement of when to prefer act versus observing, or prerequisites like an active phone session, so it stops short of full when/when-not coverage.

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