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

Annotations already declare readOnlyHint=false, destructiveHint=true, and idempotentHint=false, so the safety profile is covered structurally. The description adds genuinely useful non-schema context: the 64 KiB body limit and the dy sign convention. What it omits is the consequence side — that actions are applied to a live machine, whether they are ordered, and what happens on partial failure — which matters for a destructive tool.

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?

Four telegraphic sentences with no filler, and the most failure-prone detail (coordinate space) is front-loaded. The terseness is efficient rather than padded; the weakness is missing content, not wasted words.

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 destructive, non-idempotent tool with a deeply nested action schema, no output schema, and no per-field schema docs, the description is far too thin. It never explains the target machine identifier, action sequencing, return behavior, or error semantics, leaving the agent to guess on a high-risk call.

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 0% and the nested action object has seven fields plus a top-level 'id', so the description must compensate heavily. It clarifies x/y units, dy units and direction, and the body size ceiling, but leaves 'id', the meaning of each 'type' value, key syntax, text encoding, and button semantics entirely undocumented anywhere.

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 never states what the tool does — no verb+resource such as 'sends synthetic mouse/keyboard input to a machine'. It only supplies unit conventions (pixels, notches, 64 KiB limit), which an agent could only map to a purpose by inference from the schema's action enum. With a bare name like 'input' and no title, the tool's function is left largely unstated.

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 only guidance is 'Inspect the screen before repeating uncertain input', which implies a screenshot-first retry discipline but names no sibling tool and gives no when-to-use/when-not-to-use framing. There is no mention of when this tool is preferable to alternatives such as screenshot or ssh, so routing must be guessed.

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