windows_action

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

Annotations already declare readOnlyHint=false, destructiveHint=true, idempotentHint=false and openWorldHint=false, so the safety profile is largely covered. The description adds the re-snapshot workflow constraint, which is genuine behavioral value, but it does not indicate which of the eight actions are state-mutating (invoke/setValue) versus benign (focus/scrollIntoView), nor how failures surface when a ref goes stale.

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?

Two sentences, no filler, and the core capability plus its action set is front-loaded before the operational follow-up. Every clause 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 5-parameter, non-idempotent, destructive UI-automation tool with no output schema, the description covers the action surface and one workflow rule but omits what happens on success/failure, ref staleness behavior, and any permission or focus prerequisites. It is workable but leaves an agent to infer too much before a destructive 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 coverage is only 40%: pid and title are documented in the schema, but ref, value, and action have no field descriptions. The description partially compensates by listing the full action vocabulary and identifying windows_snapshot as where ref comes from, but it says nothing about the ref format or which actions consume value, leaving real gaps.

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?

States a clear resource (a native Windows UI Automation element from windows_snapshot) and enumerates the eight actions it can perform, which concretely defines the tool's scope. It implicitly differentiates itself from the snapshot tool by naming it as the source of the ref, though 'Act on' is a dispatcher-style verb that relies on the action list for precision.

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 instruction to re-snapshot after meaningful UI changes is a useful workflow cue, and referencing windows_snapshot as the ref source implies the correct sequence. However, it never states when to choose windows_action versus alternatives (browser_*, chrome_*, windows_screenshot) or any preconditions such as needing a valid ref from a fresh snapshot.

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