Run Windows desktop actions

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

Adds substantial behavior beyond the annotations: auto foreground-control acquisition, a visible AI-control overlay for the whole batch, continuous heartbeat, and automatic lease release on completion. These lifecycle traits are genuinely extra context the destructiveHint/openWorldHint flags don't convey.

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 front-loaded sentences: purpose first, then lifecycle guarantees, then the conditional leaseId rule. No filler or redundancy.

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?

No output schema, so the agent relies on the description for return behavior, which it partly covers via the batch lifecycle (overlay, lease release). Combined with annotations carrying the safety profile, it is mostly complete, though the action variants and the meaning of task/ttlMs remain unaddressed.

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% across 7 params, and the description only clarifies leaseId semantics. The core 'actions' structure, task, owner, ttlMs, deviceId and deviceName are left entirely to schema inference, so it adds meaning for only one parameter.

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 specific verb (Run) and resource (a batch of Windows mouse, keyboard, scroll, wait and window-activation actions), enumerating the action categories the tool covers. This is enough to distinguish it from the foreground_control_* and adb_input siblings without opening the schema.

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 implicitly routes the agent away from foreground_control_acquire by stating control is acquired automatically, and gives an explicit condition for leaseId ('only when intentionally joining a manually acquired foreground-control session'). It does not explicitly state when-not to use it or contrast with Android/process-injection siblings, keeping it below a 5.

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